Sumit Sagar

The crypto industry just witnessed a major milestone: MetaMask, the world’s most widely used self-custodial wallet, has launched its own native stablecoin — MetaMask USD ($mUSD).

This marks the first time a self-custodial wallet has issued a stablecoin, a move that could significantly reshape the stablecoin market, DeFi adoption, and the way millions of users interact with money on-chain.

The Rise of Stablecoins in Crypto

Stablecoins have become the backbone of crypto markets. They:

• Facilitate over $1 trillion in on-chain volume every month.

• Serve as the primary medium of exchange across DeFi, trading, and remittances.

• Bridge traditional finance and decentralized economies.

While USDT and USDC dominate the market with a combined market cap exceeding $150B (as of mid-2025), most stablecoins are issued by centralized companies and accessed through exchanges or custodial wallets. MetaMask’s entry changes this equation.

Why MetaMask USD ($mUSD) Matters

MetaMask isn’t just another wallet — it’s the default portal to Web3, with over 30M monthly active users globally. By embedding a wallet-native stablecoin directly into its product ecosystem, MetaMask removes layers of friction for users who want to:

• On-ramp dollars directly into their wallet.

• Swap, lend, or stake without leaving the app.

• Spend crypto like cash, thanks to the upcoming MetaMask Card (powered by Mastercard).

Key Features of $mUSD at Launch:

• Chains: Ethereum + Linea (Consensys’ EVM-equivalent L2).

• In-Wallet Utility: Onramps, swaps, transfers, and bridging.

• Real-World Payments: Expected by end of 2025 via the MetaMask Card at millions of merchants.

• Backed 1:1: By high-quality dollar-equivalent assets, with real-time transparency.

• Built with: Bridge (a Stripe company) for issuance + M0 protocol for decentralized liquidity and cross-chain interoperability.

A New Model: Wallet-Native Stablecoins

MetaMask’s approach introduces a new category of stablecoin:

1. Wallet-Native – Directly integrated into MetaMask, making it the default dollar option for millions of users.

2. Self-Custodial – Unlike exchange-issued stablecoins, $mUSD is designed to live inside non-custodial wallets, keeping users in control.

3. Composable – Built on M0’s liquidity layer for cross-chain interoperability, ensuring utility across DeFi and payments.

This combination could reduce reliance on centralized exchanges and encourage more people to onboard directly into Web3 ecosystems.

Strategic Impact on DeFi & Crypto Adoption

1. Linea DeFi Boost

By launching on Linea, Consensys’ L2, MetaMask ensures $mUSD becomes a foundational liquidity asset for its DeFi ecosystem. Lending markets, DEXs, and yield protocols on Linea will likely adopt $mUSD as a core stable asset — boosting TVL (Total Value Locked) and network activity.

2. Regulatory Alignment

The launch comes right after the U.S. GENIUS Act, which provides the first federal framework for stablecoins. This positions MetaMask to benefit from regulatory clarity and build trust with both retail and institutional users.

3. Payments Expansion

With the MetaMask Card integration, stablecoins move beyond crypto trading and into real-world commerce. This could accelerate merchant adoption and make crypto truly spendable in daily life.

4. Competitive Pressure

Other wallets and DeFi platforms may follow suit, creating their own wallet-native stablecoins. This could decentralize the stablecoin market further, reducing over-reliance on centralized issuers like Tether and Circle.

Industry Reactions

• Gal Eldar, Product Lead at MetaMask: “With MetaMask USD, users can bring their money onchain, put it to work, spend it almost anywhere, and use it like money should be used. We’re not just bringing people onchain. We’re building the reason they’ll never want to leave.”

• Luca Prosperi, CEO of M0: “The current incarnation of stablecoin technology isn’t fit for purpose. With M0 + Bridge, MetaMask gets the most interoperable and liquid stablecoin stack.”

• Zach Abrams, CEO of Bridge: “Issuing a custom stablecoin used to take more than a year. We reduced that to weeks, unlocking faster innovation.”

The Bigger Picture

MetaMask USD isn’t just another stablecoin. It’s the first wallet-native stablecoin with:

• Regulatory compliance via Bridge.

• Interoperability via M0.

• Mass adoption potential via MetaMask’s 30M+ users.

If successful, this move could change the way people enter and use Web3, making stablecoins as accessible and functional as online banking apps today.

The next wave of crypto adoption may not be driven by speculation, but by utility and MetaMask USD could be the catalyst.

Sumit Sagar

Artificial Intelligence (AI) has captured human imagination for centuries — from ancient myths of mechanical beings to today’s AI-powered apps transforming industries. The journey of AI is not just a story of technology, but of human ambition to create machines that can think, learn, and evolve.

Let’s explore how AI evolved — from its origins to the present day.

The Early Imagination: Before “AI” Had a Name

• Ancient myths & automata: As far back as 350 BCE, Greek philosopher Aristotle speculated about mechanical logic. The myth of Talos, a bronze giant automaton, showed mankind’s fascination with artificial beings.

• Early machines: In 1206, Arab inventor Al-Jazari designed mechanical automata like programmable humanoid robots.

• These were not AI as we know it, but they set the stage for humanity’s desire to create “thinking machines.”

The Birth of Modern AI: 1940s–1950s

The term Artificial Intelligence was first coined in 1956 by John McCarthy at the Dartmouth Conference — considered the birth of AI as a field.

Key milestones:

• Alan Turing (1950): Introduced the Turing Test, a benchmark for machine intelligence.

• Claude Shannon: Explored how computers could play chess.

• 1951: The first AI program, a checkers-playing machine by Christopher Strachey, ran on the Ferranti Mark I computer.

💡 Fact: In 1955, the program Logic Theorist by Allen Newell & Herbert Simon could prove mathematical theorems.

The First AI Boom: 1960s–1970s

AI began to move from theory into practice.

• MIT’s ELIZA (1966): The first chatbot, mimicking a psychotherapist.

• Shakey the Robot (1969): The world’s first general-purpose mobile robot built by SRI International.

• Governments like the US Department of Defense heavily funded AI research.

But optimism was greater than reality — computers were too slow and data was limited.

The AI Winters: 1970s–1990s

AI went through two major “winters” when funding and interest dried up.

• 1974–1980: First AI winter after researchers overpromised.

• 1980s: Expert systems (like XCON by Digital Equipment Corporation) revived interest, but soon collapsed due to high costs.

• 1987–1993: Second AI winter, as companies realized expert systems were not scalable.

💡 Fact: In 1980s, Japan’s Fifth Generation Computer Project spent $400M on AI — but achieved little lasting impact.

The AI Renaissance: 2000s–2010s

The rise of big data, faster GPUs, and cloud computing unlocked new potential.

• 2006: Geoffrey Hinton reintroduced deep learning, transforming machine learning.

• 2011: IBM’s Watson defeated human champions on Jeopardy!.

• 2012: The breakthrough AlexNet deep learning model won ImageNet competition, outperforming humans in image recognition.

• Google DeepMind (2014): Later acquired by Google, created AI that learned to play Atari games and, in 2016, defeated world Go champion Lee Sedol with AlphaGo.

💡 Data: Global investment in AI startups grew from $1.3B in 2010 to $40B by 2019 (PwC).

AI Today: 2020s–2025

The last five years have seen explosive growth in AI adoption — from research labs to everyday life.

Key Milestones:

• ChatGPT (2022) by OpenAI: Brought AI into the mainstream with over 100 million users in 2 months, making it the fastest-growing consumer app in history.

• Google DeepMind (2021): Solved 50-year-old protein-folding problem with AlphaFold, accelerating biotech research.

• Anthropic (2023) launched Claude, a safer AI assistant.

• Meta released LLaMA models, open-sourcing powerful AI.

• Microsoft invested $13B in OpenAI and integrated AI into Office and Bing.

• NVIDIA became a trillion-dollar company in 2023, powering AI training with its GPUs.

Adoption Across Industries:

• Healthcare: AI predicts diseases, aids drug discovery, and powers diagnostics.

• Finance: AI manages fraud detection, algorithmic trading, and risk management.

• Marketing: AI-driven personalization boosts engagement (think Amazon recommendations).

• Transportation: Self-driving cars (Tesla, Waymo) continue testing.

💡 Fact: According to PwC, AI could add $15.7 trillion to the global economy by 2030.

The AI Race: Companies Leading the Future

• OpenAI – ChatGPT, DALL·E, GPT-4/5.

• Google DeepMind – AlphaGo, AlphaFold, Gemini AI.

• Microsoft – Azure AI, Copilot.

• NVIDIA – Hardware powering nearly all AI models.

• Anthropic – Claude AI assistant.

• Meta – LLaMA open-source models.

• Amazon – AWS AI, Alexa.

• Apple – On-device AI for iOS.

The Future of AI: What’s Next?

• Artificial General Intelligence (AGI): Machines that can think and learn across any domain.

• AI Regulation: Governments (like the EU AI Act 2024) are setting rules to ensure safe use.

• AI + Quantum Computing: Could accelerate problem-solving exponentially.

• Everyday AI: From personal tutors to AI doctors, integration will be seamless.

💡 Fact: Gartner predicts that by 2030, 80% of all enterprise applications will have AI embedded.

Conclusion

The journey of AI is a story of cycles — excitement, setbacks, breakthroughs, and transformation. What began as a philosophical idea thousands of years ago is now shaping economies, industries, and daily life.

From Alan Turing’s vision to ChatGPT’s revolution, AI has become one of the most powerful forces of the 21st century. And while challenges remain, one thing is clear: the AI journey has only just begun.

Vikramsinh Ghatge

A manager in the buying group heard about your product in a private Slack. Someone came across a ninety second clip shared on a WhatsApp group. Another teammate skimmed your pricing page, jumped to security page, and pasted a Reddit thread. By the time a demo was booked, the room already had opinions, a shortlist, and two gotchas ready. Your funnel says awareness, interest, consideration. Their reality says decide fast with proof from people they trust. That gap explains a lot of stalled pipeline.

In Edition #57 of Vik’s M.I.X., here is what we are breaking down:

• The under 40 buyer rewrote the playbook

• LinkedIn is not your video destination

• Search traffic is shifting from pages to answers

Grab a coffee. Let’s make this practical.

The under 40 buyer rewrote the playbook

Your buyer is not waiting for a white paper. They are forming an opinion in the wild. Private chats, creator clips, internal threads, and a handful of pages on your site that either answer real questions or send them back to their peers. A large share of buying groups now sits under 45. These teams prefer to research on their own, they spend only a sliver of time with vendors, and many walk in with a preferred choice already. If you are not visible upstream with proof that feels human, you are late.

So what actually works with this buyer. Proof that travels, expertise that loads fast, and decision aids that remove risk. That is it. Notion shows how to do this without shouting. Templates, community, and creator led demos make the product feel like a friend who helps you finish a task. Atlassian turns documentation, pricing, and plan clarity into demand. Buyers can answer tough questions without booking a call. Canva floods social with short how to clips that resolve into a clean library. Discovery and activation live close together.

LinkedIn is not your video destination

No one opens LinkedIn thinking I have twenty minutes for your brand film. They open it between meetings, skim three posts, answer a DM, and leave. The feed is a crowded lobby. You pass through it. You do not camp there. So stop trying to host a screening in a hallway.

Long clips on LinkedIn are the marketing version of reading poetry on a runway. Wrong format, wrong place, wrong audience state. The sound is off. The frame is tiny. Context shifts every few seconds. The algorithm cares more about professional relevance and conversation than about letting you rack up watch time. If the point can be made with a tight sentence or a simple carousel, do that. Video is for moments where motion earns its place. A product move that needs a quick screen share. A gesture, a face, a demo that lands only when you see it.

Search traffic is shifting from pages to answers

Search still drives intent, but the shape of attention changed. More queries now serve an instant answer on top. When that summary appears, people click fewer links. Publishers and B2B content hubs are already seeing search referrals drop, often from the single digits to the mid twenties. The share of queries that trigger these summaries has grown to real scale. The biggest players are planning for fewer clicks and more on page answers. If you still measure success as page views to deep content, prepare for a shock.

This does not mean your content stops working. It means the job of your content shifts from ranking only to being findable and useful in more places. The path forward is steady.

Build owned routes. Send one weekly email that teaches a concrete lesson. Keep it short. Feature one use case and one clip. Host one community surface where honest questions get answered. Measure saves, replies, and direct landings on decision pages. These are the new early signals of intent.

Ship structured answers. Give every important page a tight TLDR block, a fast FAQ, and a short video. If a calculator or comparison table helps a buyer choose, add it. This makes your content portable. It also makes your content easier for your own team to reference in outreach.

Treat your site search like a product. Make it fast. Fix synonyms and typos. Label your best content so the right page shows up first. If buyers cannot find the answer on your site in a few seconds, they will end the session or ask a peer.

Decide your stance on AI usage. Some assets you should keep open. Some you should license. Some you should block. Tag the crown jewels so you control how they appear inside AI answers and knowledge panels. If you work with partners, make sure your most accurate descriptions and comparisons are easy to syndicate.

Rebalance measurement. Add assisted replies, saves, branded search share, and direct landings on pricing, security, and comparison pages. Clicks will move around. Intent still leaves tracks.

Run a quick diagnostic this week. Pull Search Console for your top fifty pages. Flag any page where impressions rise but clicks fall. Those pages need stronger TLDRs, richer FAQs, and short video answers that travel. Then watch if direct landings and saves rise after the change.

To conclude, Buyers are not waiting. They are deciding. Teams that win now show up where learning happens, close the gaps in distribution, and package answers that travel even when the click does not.

Sumit Sagar

📣 Today marks a seismic shift in how everyday people will access digital assets in Japan. SoftBank-backed PayPay has announced it’s buying a 40% stake in Binance’s Japanese arm — forging a deep alliance between one of Japan’s most trusted payment platforms and the largest global crypto exchange.

𝗛𝗲𝗿𝗲’𝘀 𝘄𝗵𝘆 𝘁𝗵𝗶𝘀 𝗺𝗮𝘁𝘁𝗲𝗿𝘀 — 𝘁𝗼 𝘂𝘀𝗲𝗿𝘀, 𝘁𝗼 𝗕𝗶𝗻𝗮𝗻𝗰𝗲, 𝘁𝗼 𝗦𝗼𝗳𝘁𝗕𝗮𝗻𝗸, 𝗮𝗻𝗱 𝘁𝗼 𝘁𝗵𝗲 𝗲𝗻𝘁𝗶𝗿𝗲 𝗰𝗿𝘆𝗽𝘁𝗼 𝗶𝗻𝗱𝘂𝘀𝘁𝗿𝘆:

𝟭. 𝗧𝗵𝗲 𝘀𝗰𝗮𝗹𝗲: 𝗯𝘂𝗶𝗹𝘁-𝗶𝗻 𝗱𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗶𝗼𝗻 𝗮𝘁 𝘀𝗰𝗮𝗹𝗲

PayPay has rapidly grown into Japan’s leading digital payment platform, boasting around 70 million users — that’s roughly one in every two people in the country. With over 5 million registered merchants, PayPay has built one of the largest acceptance networks in Japan’s payment ecosystem.

Dominating the QR payment space, the company holds an impressive 67% market share, underscoring its widespread adoption across consumers and businesses alike. In 2024 alone, PayPay facilitated nearly 380 million remittance transfers, reflecting its role not just in retail payments but also in peer-to-peer transactions.

Today, approximately one in every five cashless transactions in Japan is powered by PayPay — a testament to its deep integration into the nation’s shift toward a cashless economy.

When you combine that captive user + merchant base with Binance’s crypto infrastructure, you get an embedded “crypto onramp” inside a trusted payments app.

𝗜𝗻𝗶𝘁𝗶𝗮𝗹 𝗶𝗻𝘁𝗲𝗴𝗿𝗮𝘁𝗶𝗼𝗻 𝗽𝗹𝗮𝗻𝘀 𝗶𝗻𝗰𝗹𝘂𝗱𝗲:

• Buying crypto using PayPay Money (the wallet side of the PayPay ecosystem) directly on Binance Japan 

• Withdrawing crypto proceeds back into users’ PayPay wallets

This isn’t about making people jump through hoops to access crypto — it’s about making crypto a tap, not a leap.

𝟮. 𝗪𝗵𝗮𝘁 𝘁𝗵𝗶𝘀 𝗯𝗿𝗶𝗻𝗴𝘀 𝘁𝗼 𝘂𝘀𝗲𝗿𝘀 & 𝗰𝗼𝗻𝘀𝘂𝗺𝗲𝗿 𝗮𝗱𝗼𝗽𝘁𝗶𝗼𝗻

• Lower friction: Users don’t need to download separate crypto apps, set up new wallets, or learn new onboarding flows. The crypto experience is integrated into what they already use daily.

• Trust & legitimacy: Many non-crypto users hesitate due to security, complexity or trust concerns. PayPay is already a household name — “if it’s safe inside PayPay, maybe crypto is too.”

• Financial inclusion: This opens crypto access to people who might never otherwise venture into exchanges — retirees, small merchants, the less tech-savvy.

• Seamless liquidity loops: Crypto gains can feed back into fiat spending (via PayPay wallet) without friction. That’s a closed loop many exchanges can only dream of.

• Education & adoption flywheel: As more people try crypto through PayPay, word of mouth and usage will seed deeper engagement — DeFi, NFTs, stablecoins, etc.

𝟯. 𝗪𝗵𝗮𝘁 𝗶𝘁 𝗺𝗲𝗮𝗻𝘀 𝗳𝗼𝗿 𝗕𝗶𝗻𝗮𝗻𝗰𝗲 & 𝗦𝗼𝗳𝘁𝗕𝗮𝗻𝗸

For Binance:

• Guaranteed reach & volume: Rather than competing to acquire users one by one, Binance gets instant access to tens of millions of users.

• Stronger local positioning: By partnering with PayPay (a trusted Japanese brand), Binance can further embed itself into Japan’s regulatory, cultural, and financial fabric.

• Regulated compliance anchor: Binance Japan operates under Japan’s regulatory regime (registered with Kanto Local Finance Bureau). This deal fortifies Binance’s legitimacy in one of the world’s strictest markets.

• Ecosystem synergies: Together they can build new products — crypto payments, tokenized assets, yen-stablecoins, embedded finance, etc.

For SoftBank / PayPay:

• Next frontier for fintech monetization: Beyond payments rebates and commissions, PayPay now becomes a gateway into higher-margin financial services (crypto, yield, asset management).

• Defensive moat: This deal makes PayPay’s ecosystem stickier — users are less likely to stray if it’s your one portal for all financial needs.

• Valuation upside: As PayPay deepens its financial stack, its future valuation (especially if it IPOs or lists abroad) can be re-rated. (Also note: PayPay has filed for U.S listings in recent months).

• Strategic evolution: It signals SoftBank’s intent to push deeper into Web3/AI/Fintech convergence — not just backing startups, but owning key infrastructure layers.

𝟰. 𝗜𝗻𝗱𝘂𝘀𝘁𝗿𝘆 & 𝗺𝗮𝗿𝗸𝗲𝘁 𝗶𝗺𝗽𝗮𝗰𝘁 — 𝘁𝗵𝗲 𝗿𝗶𝗽𝗽𝗹𝗲𝘀 𝗮𝗿𝗲 𝗯𝗶𝗴

A. Rewriting the adoption playbook

This is a textbook case of “distribution-first” crypto adoption. Rather than waiting for users to come to crypto, you bring crypto to where users already are. If successful, we could see more payment apps worldwide integrating digital assets — the “buy crypto” button becomes just another option in your wallet.

B. Competitive disruption

Japanese incumbents (like bitFlyer, Coincheck, etc.) just got a Goliath challenger with deep capital, user reach, brand trust, and regulatory alignment. They’ll have to respond fast.

C. Regulatory momentum

Japan is reportedly preparing to give crypto assets formal status as financial products by 2026. A move like this signals to regulators that crypto isn’t fringe — it’s becoming integral to Japan’s financial infrastructure.

D. Global template & benchmarking

If Japan succeeds in merging payments + crypto securely, this becomes a model for markets like South Korea, India, Europe, etc. Countries will observe — “can we do the same in our payments super-apps?”

E. Token economics & liquidity effects

Expect deeper liquidity, more flows into onchain assets, perhaps improved usage of JPY-stablecoins, and greater integration between fiat wallets / tokenized assets. The boundary between “money” and “crypto” blurs further.

𝟱. 𝗥𝗶𝘀𝗸𝘀 & 𝗰𝗵𝗮𝗹𝗹𝗲𝗻𝗴𝗲𝘀 𝘁𝗼 𝘄𝗮𝘁𝗰𝗵 (𝗯𝗲𝗰𝗮𝘂𝘀𝗲 𝗻𝗼𝘁𝗵𝗶𝗻𝗴 𝗶𝘀 𝗳𝗿𝗶𝗰𝘁𝗶𝗼𝗻𝗹𝗲𝘀𝘀)

• Regulatory backlash or delays: Even in Japan, crypto rules are complex. Licensing, tax, consumer protection — all eyes will be on how this union is governed.

• Security & trust shocks: Any hack, fund mismanagement, or user loss could damage both PayPay’s and Binance’s reputations — more so because of the high stakes.

• User education gap: Many PayPay users may not fully grasp crypto risks, volatility, or custody. The migration must be accompanied by strong education, UI guardrails, and risk controls.

• Operational complexity: Integrating wallets, compliance, KYC/AML flows, liquidity management, settlement rails — all challenging.

• Volatility & market sentiment: If crypto markets take a hit, adoption might stall; users may lose interest or get spooked early.

𝟲. 𝗛𝗲𝗮𝗱𝗹𝗶𝗻𝗲𝘀 𝘆𝗼𝘂 𝗰𝗮𝗻 𝗱𝗿𝗼𝗽 𝗻𝗼𝘄 (𝗳𝗼𝗿 𝘃𝗶𝗿𝗮𝗹𝗶𝘁𝘆)

“Crypto in your payments app, not your crypto app”

“From cashless to tokenless: PayPay + Binance are blurring Money & Crypto”

“70 million users just got a crypto button”

“When payments become investment rails — the future of money”

“This isn’t just a crypto deal, it’s the blueprint for how Web3 will enter the mainstream. Who’s next — India? Europe? Where will the ‘crypto-button’ app emerge in your market?”

Hari TN

Suddenly everyone seems to have realized that when a service that you can’t almost live without is provided for free, then you are the product!

When man domesticated the horse, life changed all of a sudden. Long-distance travel became easy, and productivity in agriculture took a leap. It also allowed criminals to commit crimes and get away faster. It changed the way wars were fought. The story is the same when the automobile was invented. It wasn’t different when man invented the radio or learned how to split the atom.

At each of these junctures, there were intense debates on whether these inventions changed life for the better or for the worse. In the early stages, there was clamor to do away with them. In each of these cases, wisdom prevailed in the end, and societies decided not to throw the baby away with the bathwater. The inexorable march of technology continued.

You can see similar noise today when it comes to new technologies or new ways of doing business. Taxi aggregators were banned as recently as in 2014. In less than four years, this sounds so distant and absurd!

Now let’s fast forward to the debate around data privacy:

We’ll take the debate around Aadhaar and the debate around FB/Google separately.

First: Aadhaar

Aadhaar solves two long-standing problems that have crippled the nation, namely:

a) Massive leakage of benefits

Leakages occur when the subsidy does not reach the recipient due to corruption and pilferage. When well-off families, who are not the intended recipients of these benefits, received these benefits through falsification of data and documents, the intended outcomes of the program were compromised.

For example, traditionally, the bottom 50% of India’s households consumed only 25% of subsidized LPG. This meant that the subsidy on LPG was being cornered by the relatively well-off sections. Therefore, the government was really trying to fill a leaking bucket.

Similarly, more than 40% of the subsidized kerosene meant for the public distribution system was being diverted to adulterate automotive fuels. Creating a differential pricing for the underprivileged was creating havoc and ample room for pilferage.

The government’s Direct Benefits Transfer (DBT) plan, which involved transferring the subsidy amount directly to the beneficiaries’ bank accounts, began addressing this problem. The poor are now using the money transferred to their accounts to buy LPG/kerosene and other commodities at market prices.

The strategy of doing away with subsidies and transferring subsidy to deserving beneficiaries is allowing competitive markets to function for the commodities in question, besides avoiding leakage of subsidy to the non-deserving.

Leakages have come down significantly. Mint reported in May 2017 that over 23 million fake ration cards had been scrapped, potentially saving the government ₹14,000 crore in food subsidy every year. Another Mint report in August said that three states had discovered about 2,72,000 fake students availing of the mid-day meal scheme. These fake names have since been taken off.

Similarly, there were millions of fake bank accounts opened with falsified documents, and many individuals were not even aware that accounts had been opened in their names by unscrupulous individuals. Verification of all accounts by Aadhaar is cleaning up the mess.

b) A parallel economy that rivaled and emasculated the mainstream economy

The battle against the parallel economy was an unequal one in the past. Aadhaar has equipped the government with the right weaponry to kill the parallel economy.

After identifying a huge number of fake PAN cards, the government deactivated more than a million by mid-2017. The government also detected fake PAN cards allotted to non-existing individuals or people who had submitted falsified information about themselves.

With the linking of financial transactions and assets to Aadhaar, the battle against the parallel economy is truly on. There may be some friction on the ground when it comes to making these linkages, but these problems are being addressed.

Deeply entrenched vested interests are now using all kinds of surrogate concerns that sound lofty, like privacy and security.

From time to time, the scare of security breaches in the Aadhaar database has made headlines. R.S. Sharma, Chairman, TRAI, demolishes the scare-mongering by clarifying that in the seven years of Aadhaar, not a single biometric data has leaked.

The anti-Aadhaar lobby has been misusing public ignorance on what constitutes a breach to spread fake news and limit the use of Aadhaar. Sharma pulls no punches when he says that as long as Aadhaar was used by the poor for getting their entitlements, it was deemed alright. Now that it is being used by the government to check tax evasion and benami properties, it has begun to hurt some people who are using every means at their disposal to block it and limit its use.

Sharma is confident that Aadhaar is here to stay and explains that like an evolving organism, it continues to make improvements by adding new features and security layers.

While there are genuine concerns about privacy, the answer is to strengthen the security rather than throwing the baby out with the bathwater.

Now to FB/Google/Amazon et al owned data

In the early years of outsourcing, there were a couple of cases where call center agents in India working at a contact center of an outsourced service provider to a large MNC accessed credit card data of customers to pilfer money from their accounts.

There was a huge outcry, especially in the US, about the ability of Indian service providers to assure data security. The outcry finally died down because:
a) The cost of in-sourcing services back to the US was hugely expensive,
b) Even that would not guarantee data security, and
c) Indian service providers quickly scrambled to put together processes that addressed the issue.

The response to a solution or a technology that creates huge benefit with a 5% chance of misuse or fraud is to address and minimize the probability of the 5% misuse happening, rather than doing away with the technology or business model altogether — especially if the 5% chance of misuse is in the realm of possibility rather than having actually occurred.

The outcry around FB data leakage has multiple dimensions. The outrage has been amplified because the data was used to influence the US elections.

Influencing elections is nothing new. Whether attempts to influence elections and public opinion work is questionable. Much as one may dislike Trump, it is difficult to believe that the Russians were responsible for his victory. It is much more likely that he just ran a smarter campaign.

There were routine accusations in the past that the US and CIA tried influencing elections (and other events) in several countries. Similarly, the erstwhile USSR made contributions to specific individuals and parties that they thought would best protect their interests in the US.

The traditional print media too had their favorite party. Some TV channels are owned by individuals with strong party affiliations that the common man did not even know about. Hence, she was being subjected to views that she thought were neutral but were actually biased.

Every innovation in technology has amplified the ability to influence opinions by increasing the reach. Social media has simply enlarged that reach.

Social media’s huge reach has had a positive spin-off. It helped address several problems — dictatorial governments were overthrown (NYT Article), lost children found, and communities were mobilized to fight for social causes, among many others.

While discussing possible damage that social media can create, let us not forget the positives it has delivered for society. Like it or not, social media has irreversibly changed the way people interact and connect. If data privacy is taken to an extreme, then the way people connect and interact has to also go back to what it was.

A lot of the data that companies like Google have about individuals is largely used to target ads, and this service would not have been free if targeted advertising wasn’t possible because of access to data. Finally, there are very few real examples of damage caused by the big three having access to your data.

In Conclusion

Are data privacy concerns misplaced, or is there an exaggeration of the perils of leakage?

Not at all. These concerns can best be addressed if both sides of the debate are open to views and the governments of the day stop responding in a knee-jerk manner to appease angry mobs. Angry mobs are usually swayed by a few opinion makers who take extreme positions.

As long as regulation is introduced with an intention to address the 5% misuse and not lose sight of the 95% cases where there are huge benefits, it is fine. If they take a blunt approach of heavy regulation that kills the core, then we will all be wondering why we fed these imaginary fears.

Vikramsinh Ghatge

Picture this: a CIO lands on your website, casually browsing. Before they’ve even downloaded a whitepaper, AI has already mapped out their pain points, budget, and preferred tech stack.

By the time they see an ad or open an email, it doesn’t just feel relevant—it feels like it was written for them.

This isn’t a distant future. It’s happening right now.

The best B2B marketers aren’t just running campaigns anymore; they’re building intelligent systems that test, optimize, and personalize on autopilot. These self-optimizing machines remove bottlenecks, make decisions in real time, and turn static campaigns into dynamic growth engines.

In this article, we’ll explore three major shifts redefining B2B marketing:

• Self-Optimizing Systems

• AI-Driven Hyper-Personalization

• The Future of Content Repurposing

Self-Optimizing Systems: Marketing on Autopilot

Imagine running a massive B2B campaign where every ad, email, and landing page tweaks itself in real time—like a marketer with superpowers.

That’s what today’s self-optimizing systems deliver. They test, learn, and evolve at breakneck speed, ensuring your marketing never goes stale.

Real-Time Multivariate Testing

Forget A/B testing two creatives. Modern MLOps platforms can run 1,400+ creative tests across 22+ channels simultaneously.

• Example: If LinkedIn short videos suddenly get 23% more engagement in APAC mornings, the system instantly spins up dozens of new variations.

Key Components Under the Hood

• Data Ingestion Mesh: Processes millions of signals per second from CRMs, analytics, and IoT sensors.

• Evolutionary Creative Studio: Uses GANs to generate ad variations that learn from past successes.

• Allocation Optimizer: Redistributes budgets every few minutes to maximize ROI.

• Closed-Loop Learning: Retrains continuously, improving outcomes without exposing sensitive data.

Brands using these approaches have cut campaign refresh cycles from weeks to hours—boosting engagement and profitability.

AI-Driven Hyper-Personalization

Modern B2B buyers expect every touchpoint to feel tailor-made. With AI, that’s possible at scale.

Smart Customer Profiling

Neural networks now track 200+ behavioral signals in real time, from hover time on pricing pages to webinar watch duration. If a prospect lingers without converting, AI adjusts the next interaction—a discount, a case study, or peer validation—right when it’s most relevant.

Context-Aware Content Generation

NLG tools can craft multiple tailored versions of the same asset for different stakeholders.

• Example: A cybersecurity pitch generates a CFO-focused cost-saving brief, a CTO-focused technical integration guide, and a compliance officer-focused regulatory note—all in the brand’s voice.

Predictive Journey Orchestration

Instead of static nurture flows, AI dynamically adjusts channels and timing. A disengaged CTO might stop getting emails and instead see LinkedIn Thought Leader Ads—minimizing waste and maximizing resonance.

AI-Powered Content Repurposing

Content no longer ends where it starts. AI turns every asset into a content ecosystem.

How GANs Transform Marketing Assets

Generative Adversarial Networks (GANs) act as creative engines, breaking down and reassembling content in multiple formats:

• Webinars → podcasts, explainer videos, and interactive tools

• Whitepapers → decision trees and visual infographics

• Blog posts → bite-sized social content or even comic-style explainers

Case Study: Drift’s 890% ROI

By repurposing existing content with AI, Drift achieved:

• 45% higher engagement

• 32% more conversions

• 60% lower production costs

The Roadblocks

• Data quality: AI needs clean inputs.

• Brand consistency: Systems must be trained on your tone and style.

• Transparency: Always disclose AI involvement to maintain trust.

What’s Next?

The horizon for AI-powered B2B marketing is expansive:

• Next-gen GANs: Faster, more realistic, and more affordable.

• Immersive formats: AR/VR-powered experiences.

• Deeper personalization: Content tuned to individual behaviors, not just segments.

The takeaway? Marketing is no longer about campaigns—it’s about building self-learning ecosystems that adapt in real time.

The companies that master these tools won’t just keep up; they’ll dominate.

The news of Oracle laying off 10% of its India staff after the Trump meet and OpenAI deal has once again put the spotlight on the AI vs. jobs debate. The fear is real, the headlines are dramatic, but beneath it lies a hard truth we need to face.

Here’s the story and the lesson no one’s talking about:

1. If AI replaced your role, what were you really doing?

The first reaction is panic: “I’ve lost my job to a machine!” But the real question is: “Why was it replaceable in the first place?” If a line of code or a trained model can do the same work faster and cheaper, maybe it was never value-creation to begin with.

2. Great that you were fired! What better could have happened?

Losing a job to AI feels brutal, but it might be the wake-up call you needed. Better it happens now, than five years later when the gap is even wider and your skills even more outdated.

3. The illusion of “support work”

Many jobs are not about creating value, but about supporting tasks that keep the machine running without truly moving the needle. If AI could replace it, chances are, you were maintaining the illusion of work, not the impact.

4. Time to re-invest in yourself

Instead of spiraling, double down on what AI cannot easily replicate: critical thinking, creativity, strategy, leadership, empathy, design, storytelling. These are the levers of the future.

5. The best thing that could have happened to you

Yes, it stings. But this “shock” is a hidden blessing. It forces you to course-correct, learn, and bounce back stronger. Every great reinvention story begins with a disruption. This could be yours.

Jismi Jobin

Everywhere we turn today, we hear the same refrain: “AI is changing everything.” Organizations across industries are accelerating growth through AI, automation, and data intelligence. It’s an exciting time—but for many leaders, it also brings a sense of uncertainty.

• Are we doing enough to stay ahead?

• Will we be left behind in the AI revolution?

• Where do we even begin?

Having worked closely with hundreds of organizations on their transformation journeys, my advice remains consistent: don’t start with AI—start with your data.

And that foundation begins with your ERP.

Data First, Then Intelligence

For AI to deliver meaningful outcomes, your business data must be accessible, accurate, and timely. Yet, for many organizations, data remains fragmented—spread across siloed systems, outdated tools, and manual processes.

Here are some critical questions every leadership team should consider:

• Are we deriving real, actionable insights from our data?

• Is our system centralized, or are we still navigating silos?

• Are we accessing real-time data, or relying on delayed reports?

• How much time is spent explaining business logic just to extract the right data?

• Are we operating with flexibility—or constrained by legacy systems?

• Are we people-dependent (reliant on individuals) or process-driven (empowered by systems)?

• Is our infrastructure prepared for AI, or resistant to automation?

• Are our tools enabling growth—or inhibiting it?

These are not simply IT questions. They are core to business agility, resilience, and readiness for the future.

Why ERP Transformation Must Come First

At Jobin and Jismi, we have supported over 300 organizations in accelerating their digital transformation through Oracle NetSuite, a leading cloud ERP platform. The outcomes are tangible and consistent:

• A single source of truth for all business operations

• Real-time visibility across departments

• Seamless integration with third-party systems

• Elimination of fragmented spreadsheets and disconnected reports

• Immediate readiness for AI, automation, and advanced analytics

A modern ERP is not just a back-office system—it becomes the backbone of your business intelligence, operational agility, and scalability.

Lay the Right Foundation for AI

In today’s fast-paced landscape, organizations that act with clarity, speed, and precision will lead. But if your business still relies on outdated systems and disconnected workflows, you may be running blind in a data-driven world.

Remember: AI is not a magic switch—it is a multiplier. It can only enhance what already works.

Before chasing the latest technology trends, pause and reflect:

Is our foundation strong enough to support the future we’re aiming to build?

If you're seeking clarity in your transformation journey, now may be the right time to explore what modern ERP can unlock for your business.

As organizations begin to move beyond AI pilot programs, the real challenge begins — scaling responsibly. While early successes with AI pilots generate enthusiasm across teams and functions, this rapid momentum can easily spiral into fragmented efforts, unmanaged risks, and misaligned investments.

This is where the CFO must re-enter the conversation — not merely as a financial steward, but as a strategic architect of responsible AI scale.

Why Finance Must Lead the Charge

CFOs are uniquely positioned to lead AI governance — not just because they oversee financial outcomes, but because they:

• Have visibility into enterprise-wide workflows

• Are accountable for internal controls, compliance, and risk

• Influence capital deployment, resource allocation, and long-term strategy

In short, the CFO is equipped to balance ambition with accountability.

Four Pillars of Responsible AI Scaling

1. Define Clear Guardrails

As AI tools and initiatives proliferate, establish governance boundaries early:

• Identify approved use cases

• Control data access, especially for large language models (LLMs)

• Clearly assign ownership of risks

This helps prevent "shadow AI" initiatives from emerging in silos.

2. Centralize Governance, Decentralize Execution

Form a cross-functional AI Council — including Finance, Technology, Risk, and HR — to:

• Approve vendor tools and AI architectures

• Define principles such as explainability, auditability, and fairness

• Monitor enterprise-wide usage and performance

Allow individual functions to implement AI solutions, but within the established guardrails.

3. Safeguard Data Integrity

AI can only scale on a strong foundation of data. The CFO must ensure that:

• Master data is accurate and clean

• Contextual information is captured at the source (e.g., tagging journal entries, invoice comments)

• All AI activity remains traceable

AI must not only rely on quality data — it must also improve it.

4. Ensure Auditability

AI systems cannot operate as opaque "black boxes." To mitigate compliance, audit, and reputational risks, CFOs should push for:

• Transparent, explainable models

• Detailed log trails

• Clearly documented source-to-insight workflows

Scaling Is Not Just a Technology Decision — It’s a Strategic Imperative

As your organization scales AI, the CFO should be asking:

• Are we investing in AI where it delivers real business impact?

• Are we preparing our teams to work with AI, not around it?

• Are we continuously refining our models as the business evolves?

A long-term roadmap is ideal — but even a 90-day AI governance sprint can set the right tone.

Final Thought: Accelerate with Control

While piloting AI is energizing, scaling it is complex, political, and essential. It’s precisely at this stage that CFOs must step in — not to slow down innovation, but to ensure it scales sustainably and delivers lasting value.

In collaboration with CIOs and business leaders, CFOs have the opportunity to guide AI from a promising experiment to a durable enterprise asset — by accelerating what works and controlling what doesn’t.

Sumit Sagar

In the technology sector, the term “unprecedented” is often used liberally. Yet, when assessing the speed and scale at which artificial intelligence (AI) has driven global internet engagement, even that term seems insufficient.

From Silicon Valley to Sub-Saharan Africa, AI has transcended being a mere add-on to the digital ecosystem—it is fast becoming the internet’s new operating system. The supporting data is nothing short of extraordinary.

A Timeline Rewritten: AI Adoption vs. Internet Growth

The global internet revolution took approximately 23 years to achieve 90% user penetration outside North America. In contrast, AI—led prominently by OpenAI’s ChatGPT—accomplished this milestone in just 3 years.

Let that comparison settle in:

📈 User share outside North America:

• Internet: 90% in 23 years

• ChatGPT App: 90% in just 3 years
  (Source: Sensor Tower & ITU, May 2025)

This explosive adoption was made possible by the digital infrastructure built over the last two decades—smartphone ubiquity, mobile-first design, and globally accessible app ecosystems all played critical roles.

A Global Phenomenon, Not a Regional Trend

AI’s proliferation has not been limited to the West. It rapidly evolved into a global utility. By April 2025, the ChatGPT mobile app boasted:

• Over 800 million weekly active users

• Significant adoption across Europe, South Asia, Latin America, Sub-Saharan Africa, and the Middle East

This wasn’t a centralized, top-down initiative—it was a bottom-up, organic global surge in usage.

Monthly Active Users by Region (as of April 2025):

• East Asia & Pacific: 27%

• Europe & Central Asia: 19%

• Latin America & Caribbean: 14%

• Middle East & North Africa: 12%

• Sub-Saharan Africa: 11%

• South Asia: 10%

• North America: 7%

100 Million Users in Days, Not Years

ChatGPT reached 1 million users in just 5 days—the fastest adoption rate in consumer tech history. To put this in context:

• iPhone: 74 days

• Netflix: Over 3 years

• Ford Model T: 2,500 days

And the acquisition cost? Effectively $0—thanks to the confluence of digital readiness and exceptional product-market fit.

The Catalysts Behind This Growth

Three primary enablers made this exponential growth possible:

1. Mobile-first accessibility – The ChatGPT mobile app brought AI directly into the hands of billions.

2. Global digital literacy – With over 5.5 billion connected users, the digital audience was primed for adoption.

3. Simplicity of user experience – The intuitive interface of large language models made the technology feel both powerful and approachable.

The Momentum Continues

This growth trajectory shows no signs of slowing. ChatGPT and similar platforms are evolving into agentic systems—capable of acting as copilots in domains such as creativity, productivity, research, and customer service.

Moreover, as AI becomes deeply embedded in everything from enterprise platforms to consumer search, user adoption is expected to surge further.

With global powers like China and the United States investing heavily in sovereign AI infrastructure, the race is as much geopolitical as it is technological.

Conclusion: A New Internet Moment

What the internet achieved over three decades, AI has replicated—and in some cases, surpassed—in just three years.

We are witnessing a technological adoption curve that outpaces every major innovation in modern history. This is no longer just about chatbots. It’s about billions of individuals gaining access to smarter, more capable digital tools—ushering in a new era of intelligence, connectivity, and opportunity.

And we are only in Year 3.

Jismi Jobin

As a business scales, choosing the right platform to manage finances, operations, and strategic growth becomes critical. The common dilemma many organizations face is whether to invest in a custom-built solution tailored to their current processes or adopt a proven, cloud-based ERP platform like Oracle NetSuite.

While custom software may initially appear cost-effective and flexible, it often becomes a major constraint as businesses grow. For companies generating over $20 million in annual revenue, outgrowing a custom-built system isn’t just possible — it’s highly probable.

The Hidden Pitfalls of Custom-Built Systems

Custom software is typically designed to mirror how your business operates today — not how it will operate tomorrow. As business needs evolve, these systems often struggle to keep pace. Common issues include:

• Architectural Limitations: Even minor changes can require extensive redevelopment. Over time, the system becomes unwieldy and complex to manage.

• Developer Dependency & Lack of Documentation: When the original developer moves on, new developers are left to decipher undocumented code and business logic, making updates slow, error-prone, and expensive.

• Outdated Technology: Custom systems may rely on legacy platforms that are no longer supported or compatible with modern tools, complicating upgrades and integrations.

• Version Lock: Custom features can break during upgrades, forcing businesses to stay on outdated versions and miss critical improvements.

These issues often result in business stagnation — not due to a lack of vision or opportunity, but because the underlying systems are unable to support new initiatives.

Oracle NetSuite: A Better Business Decision

Oracle NetSuite is a comprehensive cloud ERP platform built to scale with your organization. It integrates core business functions including finance, operations, inventory, sales, and analytics into a single, unified system.

Rather than building from scratch, NetSuite enables companies to tailor proven industry best practices to fit their needs, allowing for faster results, lower risk, and greater long-term agility.

ERP Is an Investment — Not an Expense

When treated as a strategic investment rather than a technical expense, ERP platforms like Oracle NetSuite can accelerate business growth significantly. In fact, companies leveraging cloud ERP platforms often grow up to four times faster than those depending on rigid, custom-built systems.

With NetSuite, businesses can make faster decisions, automate core processes, and scale operations across multiple locations, entities, or service lines with ease.

Why the Right Implementation Partner Matters

The success of an ERP implementation lies not just in the software’s capabilities, but in how effectively it is deployed and adopted. This is where the role of a trusted implementation partner becomes critical.

A capable NetSuite partner will:

• Gain a deep understanding of your business goals and processes

• Configure NetSuite to align with day-to-day operations

• Guide teams in utilizing standard features efficiently

• Avoid unnecessary customizations that complicate long-term maintenance

• Offer training and change management support to ensure smooth adoption

This approach ensures that you build on a strong, standardized foundation — one that enhances scalability without reinventing the wheel.

Key Advantages of Oracle NetSuite

• Flexible Configuration: Adapt workflows, roles, and approval processes without heavy development

• Scalable Architecture: Seamlessly manage multi-location, multi-currency, and multi-entity operations

• Seamless Integration: Connect easily with third-party tools and platforms

• Real-Time Insights: Access dashboards and analytics that support smarter, data-driven decisions

• Continuous Innovation: Benefit from regular updates, enterprise-grade support, and a global network

Final Thought

Your systems should empower your growth — not hinder it.

Oracle NetSuite offers the visibility, efficiency, and flexibility needed to support modern businesses in their growth journey. Whether you’re scaling across geographies or optimizing internal operations, NetSuite provides the infrastructure to grow faster, smarter, and with confidence.

Let’s transform your operations — the smart way, with Oracle NetSuite.

Sumit Sagar

Artificial Intelligence is evolving faster than any previous technological wave. But as the tide rises, it brings a surprising paradox: while the cost to train state-of-the-art AI models is skyrocketing, the cost to use them is plummeting.

This shift is catalyzing a new era where performance is converging across leading models and developer adoption is soaring—despite high upfront investments.

Let’s break down what’s happening and why it matters.

1. Training AI Models Is Expensive—And Getting Pricier

Training the latest AI models requires immense compute power. Between 2015 and 2025, the computational cost (measured in FLOPs) to train key large language models has grown at an astonishing 360% CAGR (Compound Annual Growth Rate).

• Training GPT-class models now consumes billions of dollars in compute resources.

• In 2024 alone, OpenAI’s compute costs outpaced its revenue by nearly $5 billion, according to estimates.

• Supercomputers built for AI have scaled rapidly, with performance growing at 150% annually, driven by denser chip clusters and better hardware.

This AI "arms race" is largely dominated by a handful of major players with deep pockets: OpenAI, Google, Meta, xAI, Anthropic, and their Chinese counterparts.

2. Inference Costs per Token Are Dropping Fast

While training models is becoming more expensive, the cost of using them—known as inference—is falling sharply.

• Inference costs per token have declined steeply, mirroring the historical cost curves of electricity and memory.

• Just as memory and energy became exponentially cheaper over decades, AI inference is now following a similar deflationary trajectory.

This makes it dramatically more affordable to build and deploy applications on top of powerful models—without having to train them from scratch.

3. The Net Result: Performance Is Converging

As more models reach massive scale and inference becomes commoditized, we’re seeing a phenomenon the report describes as:

“Performance converging + developer usage rising.”

What this means:

• Leading models are achieving comparable performance on key benchmarks like MMLU and human-level reasoning.

• Open-source models like LLaMA 3, Qwen 2.5, and Grok 3 are catching up with proprietary ones in real-world use cases.

• With marginal gains from larger models shrinking, fine-tuning and use-case-specific customization are becoming more valuable than raw scale.

In short: it’s no longer just about who has the biggest model—it’s about who can deliver performance efficiently and creatively.

4. Developer Adoption Is Booming

Thanks to falling inference costs and growing performance parity, the developer ecosystem is exploding:

• NVIDIA’s AI developer base has grown 6x to over 6 million in just seven years.

• Google’s Gemini ecosystem grew from 1.4 million to 7 million developers in a single year—a 5x jump.

• OpenAI’s ChatGPT app reached 800 million weekly active users by April 2025—just 17 months after launch.

This growth isn't limited to tech giants. It's spreading across startups, enterprises, and governments, embedding AI into every layer of digital infrastructure.

5. Why This Matters

This changing cost dynamic is reshaping the economics of AI:

• Startups no longer need to train massive models; they can fine-tune open-source models or use cloud APIs.

• Inference-as-a-Service is becoming the norm, allowing businesses to plug into world-class AI at a fraction of previous costs.

• Wider access = more innovation—we’re witnessing a democratization of AI development reminiscent of the open-source software revolution.

Final Thoughts

The future of AI isn’t just defined by billion-dollar GPUs or elite research labs. It’s increasingly shaped by a global community of developers, empowered by falling costs and rising capabilities.

Yes, training models is more expensive than ever—but using them is cheaper, faster, and more widespread than anyone imagined a few years ago.

The paradox of AI’s economics isn’t a bug—it’s a feature. And it's unlocking the next great wave of global innovation.

Once a buzzword for decentralization and disruption, Web3 is now carving a path toward becoming a foundational layer of the next version of the internet. As we step deeper into 2025, it’s clear that the space is maturing rapidly. From utility-focused tokens and AI integrations to enterprise NFTs and decentralized social platforms — Web3 is evolving in ways that go far beyond speculation.

So, where exactly are we headed?
Let’s explore the top trends shaping the Web3 landscape in 2025 and beyond.

1. The Rise of Utility-Driven Tokens

Data Point:
In 2024, over 63% of newly launched tokens had a clear utility use case — a 40% increase from 2022.
Source: Messari, TokenInsight Inc.

Real-Life Example:
Arbitrum’s ARB token stood out with a strong emphasis on governance and ecosystem expansion. It wasn’t about speculation — users had reasons to hold and use the token meaningfully.

Takeaway:
Web3 projects need to communicate current, tangible value. Future promises are no longer enough.

2. The Real Winners Will Be the Ones Who Onboard the Next Billion

Data Point:
Web3 users globally reached 420 million in 2024, but 80% are concentrated in just 10 countries.
Source: Chainalysis, Dune Analytics

Real-Life Example:
Worldcoin, though controversial, demonstrated the power of onboarding at scale through biometric incentives and wide-reaching campaigns.

Takeaway:
The Web3 onboarding experience must evolve to be more intuitive, accessible, and mobile-first. Crypto-native UX alone won’t scale.

3. Regulatory Clarity is Fuel, Not Fire

Data Point:
Countries with clearer regulations (e.g., UAE, Singapore) saw a 2x increase in crypto-related startup registrations in 2024.
Source: Electric Capital Developer Report

Real-Life Example:
Ripple’s partial legal win in the U.S. helped restore investor confidence and catalyzed institutional partnerships.

Takeaway:
Transparency, compliance, and open communication with regulators can give Web3 projects a significant edge.

4. Interoperability is the New Scaling

Data Point:
40% of dApps now support cross-chain functionality, up from just 13% in 2022.
Source: DefiLlama, Alchemy

Real-Life Example:
LayerZero and Wormhole are enabling seamless cross-chain communication. Apps built on these are seeing wider adoption due to reduced friction.

Takeaway:
Projects that integrate across ecosystems are more likely to scale sustainably.

5. ZK and AI Will Be the Web3 Superpowers

Data Point:
ZK-related projects received over $700M in funding in 2024, while AI-Web3 integrations grew by 240%.
Source: Crunchbase, Delphi Digital

Real-Life Example:
Polygon Labs’s zkEVM is enabling Ethereum compatibility with zero-knowledge proofs, supporting a growing ecosystem of dApps.
Bittensor is building a decentralized network of AI models — creating a token-incentivized, open-source AI layer.

Takeaway:
ZK makes Web3 scalable and private; AI makes it intelligent and adaptive. Together, they unlock the next frontier of decentralized infrastructure.

6. NFTs Are Quietly Entering Their Enterprise Era

Data Point:
68% of NFT volume in Q1 2025 came from ticketing, loyalty, or IP licensing — not art or PFPs.
Source: DappRadar, Nansen

Real-Life Example:
Nike’s .SWOOSH and Starbucks Odyssey show how NFTs can power loyalty, access, and community engagement.

Takeaway:
Think of NFTs as programmable brand touchpoints, not just JPEGs.

7. The Best Web3 Brands Will Feel Like Web2.5

Data Point:
Projects with simplified UX and fiat on-ramps saw 3.2x user retention over those requiring MetaMask or seed phrases.
Source: Coinbase Cloud, Web3 Index

Real-Life Example:
Phantom wallet’s integration of Apple Pay and Google login significantly increased daily active users.

Takeaway:
To drive mainstream adoption, simplicity will win. Remove friction, not features.

8. Decentralized Social Will Explode, But Slowly

Data Point:
Lens Protocol and Farcaster combined reached 1.2 million users by early 2025.
Source: Dune Analytics, Warpcast data

Real-Life Example:
Farcaster’s Frames feature created early viral loops. Lens is building a growing creator economy.

Takeaway:
There’s a long-term opportunity here for builders and creators. Early adoption and experimentation are key.

9. Token-Gated Experiences Will Replace Ads

Data Point:
Over 10% of major Web3 projects now offer token-based access to content or events.
Source: Bankless, Forefront

Real-Life Example:
Bankless DAO’s premium newsletter is only accessible via tokens — with over 10,000 paying members.

Takeaway:
Forget impressions — start building communities around exclusivity and earned access.

10. AI + Crypto + Creator Economy = The Trillion-Dollar Trifecta

Data Point:
Platforms like UGCx, Story Protocol, and Mirror are combining AI, tokens, and IP licensing to reshape how content is created and monetized.
Source: Variant Fund, a16z Speedrun

Real-Life Example:
Story Protocol raised $54M to help creators tokenize, license, and protect their content with AI-powered attribution.

Takeaway:
The future of media and content creation lies at the intersection of AI, tokens, and ownership.

Conclusion

Web3 is no longer a speculative playground — it’s a serious, evolving layer of the global digital infrastructure. Its future lies in utility over hype, inclusion over exclusivity, and experiences over expectations.

The projects that win will be those that blend cutting-edge tech with real human value — making Web3 not just decentralized, but desirable.

Asha Pillai

In today’s flood of AI breakthroughs, many CFOs find themselves asking quietly: Am I behind? Should I already know more?

The pace of innovation can feel relentless. Every week brings a new platform, a new acronym, a new vendor pitch. But here’s the truth: you don’t need to know everything about AI — you just need to know enough to lead with clarity.

Below is a practical framework for CFOs who want to understand and leverage AI, without getting lost in the noise.

Step One: Learn Just Enough to Ask Better Questions

You don’t need to understand the technical mechanics of LLMs, transformers, or diffusion models. What you do need is a working grasp of what AI can actually do — especially in the context of finance, operations, and enterprise value.

Today’s enterprise-grade AI tools can generate dashboards from raw data using natural-language queries, detect anomalies and patterns that impact revenue or margins, automate workflows like reconciliations and reporting prep, and act as a querying layer across complex ERP environments — giving leadership real-time access to insights, not just monthly reports.

This isn’t about cosmetic upgrades to dashboards. It’s about enabling faster, more informed decision-making.

To get started, ask yourself:

• Are our reports helping teams take better action, or are insights buried in static PDFs?

• Are we surfacing key trends when they matter most, or only in retrospectives?

• If we had an AI chatbot querying our ERP tomorrow, would it know what to look for?

These questions lead naturally to one thing: your data foundation.

Step Two: Push for Data Readiness

AI systems are only as good as the data they’re built on. That’s where the CFO plays a pivotal role.

To unlock value from AI, finance teams must ensure:

• High-quality data capture at the source, especially context-rich journal entries that explain the “why” behind transactions.

• Real-time variance analysis capabilities, supported by models that understand context, not just numbers.

• Data integration across multiple ERPs, using AI data lakes or federated approaches.

• Structured, standardized, and annotated data, so AI doesn’t just summarize the past but can drive action.

Think of it this way: you wouldn’t deploy ERP without strong chart-of-accounts discipline. Similarly, you can’t scale AI without robust data governance.

Step Three: Build a Cross-Functional AI Council

AI adoption isn’t a solo initiative. The most forward-thinking CFOs are assembling small, agile AI councils made up of finance leaders, technology teams, HR, operations, and risk or compliance stakeholders.

This council’s role is to:

• Prioritize high-impact areas where AI can drive measurable results.

• Run short, controlled pilots that demonstrate value before scaling.

• Ensure ethical and regulatory guardrails are in place from day one.

Think of it as your ERP task force — reimagined for speed, experimentation, and cross-functional collaboration.

Step Four: Anchor Learning to Business Outcomes

Avoid falling into the trap of chasing use cases in isolation. Instead, ground every AI exploration in a business outcome.

The shift in mindset is key: don’t ask “What can this AI tool do?” — ask “What decision does this unlock? What measurable impact does it create?”

For example:

• Using AI for margin analysis should help you pinpoint drops at the SKU or plant level, protecting profitability.

• Predictive cash flow forecasting should drive better working capital decisions and liquidity management.

• Tagging expense anomalies automatically should lead to faster, deeper cost investigations.

• A reporting chatbot should surface trends in real time, enabling agile leadership responses.

This is how you avoid shiny-object syndrome — and stay focused on business value.

Step Five: Set Realistic Expectations Internally

AI maturity will vary across teams. Some departments will be early adopters. Others may be slower to engage.

As CFO, your tone matters. Set the pace with calm, strategic leadership. Emphasize that your focus is on outcomes — not novelty.

Avoid two extremes:

• The FOMO trap: buying everything that looks new and exciting.

• The paralysis trap: doing nothing because the landscape feels overwhelming.

Instead, establish a clear cycle: pilot → learn → scale, backed by success metrics and internal feedback loops.

Bonus: Stay Curious — Even If You’re Not the CIO

You don’t need to be technical, but you do need to stay curious.

This means:

• Keeping an eye on AI agents and automation platforms — particularly if you run shared services or finance operations.

• Staying informed about emerging tools like Gamma (AI-generated presentations), HeyGen (avatar videos), or Microsoft Copilot (productivity automation).

• Exploring new tools on your personal laptop — safely and freely — to build your own sense of what’s possible.

The goal isn’t to become a developer. It’s to sharpen your ability to identify opportunity — and apply it where it matters.

Final Thought: Lead AI Like You Led ERP

You didn’t need to code to lead an ERP rollout. You needed to ask the right questions, align teams, and drive transformation.

AI is no different.

You’re not expected to master the models. Your role is to understand the levers, connect them to strategy, and lead with clarity, curiosity, and control.

Coming Up Next

“The CFO’s AI Playbook: Five Use Cases You Can Pilot in the Next 60 Days”

In an era defined by short attention spans and information overload, capturing mindshare in crypto is a formidable challenge. Projects are no longer just competing with other protocols — they’re competing with breaking news, viral memes, and internet culture, all in the same user scroll.

Within this landscape, storytelling is not a luxury — it is a necessity. It is the primary lever for trust, clarity, and conversion. This blog explores actionable storytelling frameworks used by leading crypto teams to create narrative-driven content that consistently generated engagement, virality, and signups — even within the most skeptical communities on Crypto Twitter, Discord, and Telegram.

1. Stories Over Features: Reframing the Value Proposition

Technical specifications rarely move the needle. Users respond to transformation, not terminology.

The Framework:

• Hero: The user

• Problem: The friction in the crypto experience (e.g., gas fees, poor UX, scams)

• Journey: How the project resolves it

• Victory: Tangible benefit to the user — whether in profit, safety, or early access

Example:

• Before: “An L2 chain with 500 TPS.”

• After: “What if gasless trading felt like sending a DM?”

By narrating around the user's journey, engagement and relatability increased significantly.

2. The “5S” Hook Framework: Five Angles That Drive Clicks

After testing multiple hooks, five formats repeatedly outperformed others in driving attention and action:

• Shock: “80% of wallets are inactive. Yours too?”

• Secret: “What we learned after losing $50K in gas fees”

• Struggle: “I gave up on DeFi… until this changed everything”

• Status: “What the top 1% of NFT collectors do differently”

• Stats: “Our DAO grew 400% in 60 days. Here’s how”

These hooks were repurposed across Twitter threads, carousel posts, and Discord announcement formats — consistently resulting in a 5x to 10x lift in engagement.

3. Familiar Formats for Novel Concepts

New ideas can confuse audiences. The solution? Embed them within familiar storytelling structures:

• “Before vs After” — Ideal for product upgrades

• “3 Mistakes Every ___ Makes” — Effective for onboarding education

• “X vs Y” — For simplifying competitive differentiation

• “How We Went From ___ to ___” — Perfect for growth milestones

Example:

“How we went from 10K to 80K users with $0 ad spend” → This post alone drove 8,000+ new site visits in 24 hours.

4. Community as Co-Creators of the Narrative

Engagement is no longer top-down. Teams that invited users to shape the story saw significantly higher retention and virality.

Strategies That Worked:

• Enabling users to vote on product or feature names

• Turning user-submitted bugs into memes

• Featuring active wallet holders as “Builders of the Month”

When users become part of the narrative, they transition from passive observers to active evangelists.

5. Repurposing High-Performing Stories Across Channels

Every winning narrative was recycled — not just reposted.

Applications:

• High-performing Twitter threads became reels

• Viral memes turned into carousel posts

• Testimonials evolved into onboarding flows

• Insights fueled investor decks and Telegram pinned posts

One strong story often became 10+ growth assets.

Conclusion

Attention is a volatile currency in the crypto world. Teams that build, launch, and ship without crafting a compelling story risk being forgotten in the scroll. Those that succeed do not just explain their product — they build a narrative that invites, inspires, and converts.

A few months ago, I was on a late-night call with a sales lead.

It was one of those pivotal moments. The quarter was on the line, deals were stalling, and the frustration in their voice—usually steady—was unmistakable.

“Shruti, I don’t know what story to tell anymore.
Product is talking value. Marketing is focused on features.
And I’m stuck in the middle.”

That comment stayed with me.

Because I’ve been that marketer. That product stakeholder. That go-to-market (GTM) owner who meticulously crafted a strategy deck—only to watch it falter under the real-time pressure of execution.

And it forced me to reckon with a tough truth:

A GTM strategy that cannot survive sales calls, shifting priorities, or customer feedback is not a strategy. It’s theatre.

That moment didn’t spark a brainstorming session. It triggered a transformation in how I thought about go-to-market execution. Not as a static launch plan, but as a dynamic operating system.

From Beautiful Plans to Brittle Execution

Just three weeks after a major product launch, the early warning signs became painfully clear:

• Sales was telling a different story than what marketing had prepared.

• Customer success was unclear on which segments to prioritize.

• Paid campaigns were generating MQLs—but very few were converting.

I revisited our GTM strategy deck. It was beautifully constructed—20 slides, many stakeholders involved, hours of thoughtful design.

But it was also static. And dangerously brittle.

This wasn’t a GTM strategy.
It was a snapshot.
Visually compelling, but structurally fragile.

What We Needed Was a GTM Nervous System

Over the last eight years, building GTM motions across AI-first SaaS, enterprise fintech, and platform-led growth environments, one pattern has emerged consistently:

Great companies don’t just launch.
They operate. In-market. With fluidity, feedback loops, and focused alignment.

Unfortunately, most GTM efforts fail not due to lack of effort, but because of:

• Over-engineered pre-launch plans

• Misaligned post-launch execution

• Siloed insights from sales, product, marketing, and customer success

• Endless confusion around ownership and accountability

What we needed wasn’t another slide deck or roadmap.

We needed a GTM Operating System—a shared system that aligns, adapts, and evolves as the market does.

Discovering the GTM OS Framework

My first introduction to the term “GTM Operating System” came from GTM Partners’ newsletter—a highly recommended read. Far from being just another buzzword, it articulated what many of us had felt but struggled to define:

“Strategy is not a one-time plan. It’s a system of decisions, metrics, and motions that lives and breathes across every team—every day.”

This concept gave structure to something I had intuitively sensed: that a static launch plan cannot meet the dynamic demands of modern GTM execution.

Supporting Data: Why Traditional GTM Fails

The numbers underscore the urgency of rethinking how we go to market:

• 95% of product launches fail to meet objectives due to poor cross-functional alignment. (Harvard Business Review, 2022)

• Teams using shared GTM metrics like LTV:CAC, NRR, and Time-to-Value are 2.5x more likely to achieve revenue goals. (Gainsight Pulse, 2023)

• Only 14% of marketers feel confident adapting campaigns based on real-time signals. (HubSpot, 2024)

• 48% of B2B companies still define GTM success by MQLs alone—not revenue, retention, or expansion.

It’s no wonder teams feel like they’re spinning their wheels, despite having well-designed plans.

What a Functional GTM OS Looks Like

Here’s how I define a working GTM Operating System based on what’s worked—and what hasn’t—in real execution environments:

1. Shared Rituals, Not Just Shared Documents

• Weekly cross-functional syncs (Product, Sales, Marketing, CS)

• Daily Slack threads on active deals and campaign responsiveness

2. Real Metrics That Reflect Reality

• Focus on Net Revenue Retention (NRR), LTV:CAC, and Time-to-First-Value—not just MQLs

• Use of tools like Gong, Wynter, and Apollo AI to capture buyer sentiment and resonance

3. Live Feedback Loops

• Integrate AI-powered sentiment analysis into campaign dashboards

• Maintain living GTM docs (via Notion, Airtable, or similar) that tie product changes to GTM readiness

4. Purposeful GTM Motions (PLG ≠ Default)

• Example: One PLG motion resulted in 4x sign-ups—but only after we fixed friction in value messaging

• In another scenario, a sales-led motion outperformed inbound by 30% because our ICP wasn’t suited for self-serve onboarding

The Emotional Reality of GTM

Here’s something we don’t talk about enough:

Most GTM breakdowns are not due to bad planning—they’re due to emotional misalignment.

Team members often feel unseen, unheard, or uncertain. That’s why a strong GTM OS should also foster:

• Psychological safety for open, cross-functional dialogue

• Time for strategic alignment—not just surface-level syncs

• Clear ownership, so responsibilities don’t fall through the cracks

The most effective GTM leaders I’ve worked with don’t just create plans.
They create clarity and conviction. Across functions. Across time. In-market.

What’s Your GTM Motion Right Now?

Are you doubling down on Product-Led Growth?
Exploring community or partner-led models?
Or perhaps still iterating?

That’s more than okay.

I’d love to hear how you’re navigating your go-to-market efforts—successes, struggles, and everything in between.

Feel free to share your thoughts in the comments or reach out directly. Let’s continue building a shared language and operational rhythm around GTM—not as an event, but as an evolving system.

Because the future of go-to-market isn’t about how well we launch.
It’s about how effectively we operate once we do.

Sandeep K Nagpal

The Rise of Artificial Intelligence (AI) Assistants: A Paradigm Shift for Marketing

The rise of artificial intelligence (AI) assistants is not just a technological advancement—it’s a paradigm shift for the marketing industry. With tools like ChatGPT, Google Gemini, and Meta AI becoming integral to how consumers search, shop, and interact, the ripple effects will fundamentally alter digital marketing strategies, budgets, and metrics.

This transformation presents both challenges and opportunities, particularly in areas like paid media, SEO, Return on Ad Spend (ROAS), and even physical formats like events. Here’s an exploration of what’s to come and how marketers can navigate this new frontier.

The Decline (and Evolution) of Paid Media Budgets

AI assistants act as intermediaries, delivering curated responses and recommendations directly to users. This reduces the dependency on traditional digital advertising channels like search and display ads. As a result, paid media budgets are likely to shrink—but this is not a simple reduction. Instead, it signals a strategic reallocation:

Ultimately, while traditional digital ad spends will decline, these budgets will evolve to support AI-driven discovery and personalized interactions.

SEO’s Reinvention in the Age of AI

AI assistants are disrupting the conventional SEO model. Instead of presenting a ranked list of search results, AI tools deliver direct answers or summaries, bypassing traditional click-through mechanisms.

Marketers will need to develop AI-first content strategies, crafting content that is concise, accurate, and contextual to secure their place in AI-generated results.

Redefining ROAS for the AI Era

Return on Ad Spend (ROAS), traditionally calculated based on direct attribution models, will need a fundamental redefinition:

• Shift from Clicks to Engagement: With AI summarizing content and reducing direct clicks, marketers must look beyond immediate conversions. New metrics might include voice-assistant engagement rates or recommendation influence.

• Focus on Lifetime Value: AI assistants emphasize relevance and personalization. ROAS will evolve to prioritize customer lifetime value (CLTV) and other metrics tied to long-term engagement and retention.

This new paradigm will push marketers to focus on the quality of interactions, not just the quantity of clicks or leads generated.

The Ripple Effects on Physical Formats Like Events

While AI is redefining digital marketing, it’s also enhancing the role of physical events in the marketing mix. Here’s how:

• Physical Events Gain Strategic Importance
  With digital interactions becoming increasingly mediated by AI, physical events offer a human touch that cannot be replicated. They allow brands to foster authentic relationships, build trust, and create memorable experiences.

• AI-Powered Event Personalization
  AI will enhance the event experience, making it more impactful and efficient:

• Hybrid Models Take Center Stage
  AI-powered virtual components will scale physical events beyond geographical boundaries, allowing brands to reach wider audiences while maintaining the intimacy of in-person interactions.

Rebalancing Digital and Physical Marketing Budgets

As AI reshapes the marketing landscape, marketers will need to strategically reallocate resources to thrive in this new ecosystem:

• Digital Spending Trends: A projected 10–15% decline in traditional digital ad budgets, counterbalanced by increased spending on AI optimization and conversational advertising formats.

• Physical Formats: Events will absorb a larger share of marketing budgets, with spending potentially increasing by 10–15%, driven by their ability to create high-touch, human-centered interactions.

Key Takeaways for Marketers

• Invest in AI Expertise: Upskill teams to understand and leverage AI-driven search algorithms and conversational ad formats.

• Prioritize AI-First Strategies: Ensure your content and campaigns are optimized for visibility in AI systems, focusing on natural language, semantic relevance, and personalization.

• Elevate Event Experiences: Use AI to personalize, scale, and measure event performance, ensuring maximum ROI from physical formats.

• Adopt New Metrics: Move beyond traditional KPIs like click-through rates or basic ROAS and embrace more nuanced, long-term metrics like CLTV and AI engagement.

Conclusion: Thriving in the AI-Driven Marketing Landscape

The rise of AI assistants represents a pivotal moment for the marketing industry. Digital spends, SEO strategies, and event budgets will all undergo significant transformations as businesses adapt to a world where AI drives discovery and engagement.

Marketers who embrace these changes, invest in innovation, and balance digital with physical experiences will not only stay relevant but also thrive in this rapidly evolving ecosystem. The future of marketing isn’t just digital—it’s AI-empowered and human-centric.

Separating AI Hype from Reality

I have no doubt that AI is making huge strides and has been occupying some of the best minds of this century. However, the hype around AI is also making equally huge strides. Some of the hype, one can argue, is inevitable—but one needs to be a little watchful because it is entering our everyday lives. Many of us are in positions, both in our professional as well as personal lives, where we need to evaluate the genuineness of claims about using AI. And if you can’t separate the hype from the truth, you may end up spending money on fake products and services.

They say that when it comes to capital markets, you are in the midst of a bubble if your driver begins to make stock recommendations to you. We can see the equivalent happening in AI now.

Let me first take a couple of examples of genuine AI products.

Some years back, one of the co-founders of Liv.ai, a Bengaluru-based AI start-up, met me and demonstrated their product that used NLP to convert speech to text in multiple Indian languages. I had always known that text-to-speech was easy, but converting speech to text in multiple languages was a hard problem to solve. I was a bit skeptical—more so because this was an Indian company. But when I saw the product, I was quite blown away. Before I could even seriously think of recommending it to someone who would see this start-up as a great investment opportunity, Flipkart acquired it and built their shopping assistant, ‘Saathi’, with a text and voice interface to support shoppers in the smaller towns of India.

Facial recognition is another problem that has been solved and already has wide applications that touch everyday lives, including unlocking one’s smartphone. Work is in progress with image recognition applications in other fields, including in horticulture.

Let me now take some examples of what I would call fake products riding on the AI wave.

A vendor once approached us claiming they had a test that could predict criminal tendency in an individual with an accuracy of 60%, and maybe we should consider this tool to evaluate our delivery boys! Very tempting for a starry-eyed, gullible user sold on AI.

Let’s dig a little deeper and subject this claim to a rigorous test. The test has a 40% error rate, which means that there is a 40% probability that it would classify someone with no criminal tendency as one with a criminal tendency. Let’s now assume that the prevalence of criminal tendency in society is 2%. So, in a population of 100, you would have 98 people without a criminal tendency. But the test would claim that 41 people had criminal tendencies (40% error rate on 98 is 39; add the 2 real ones and you have 41). So, the error rate is actually 39 on 41, or 95%. The accuracy of the test with this additional data point is just 5%!

Do you need anything else to decide whether you should pay this vendor and run all your new hires through a test like this?

Sometime in April 2019, in an interview with CNBC, Ginni Rometty claimed that IBM artificial intelligence can predict with 95% accuracy which workers are about to quit their jobs. Rometty would only say that its success comes through analyzing many data points. She would not explain “the secret sauce” that allowed the AI to work so effectively in identifying workers about to jump and just said that its success comes through analyzing many data points. IBM is offering this tool to external clients! In my opinion, this claim is totally untrue and is driven by a vested interest.

Those familiar with the butterfly effect and chaos theory would understand this well. Some systems are not very amenable to predictions for the simple reason that even the minutest variation in the initial conditions can result in a huge variation in the end results. In other words, a very small variation in the initial conditions does not result in a very small variation in the outcome!

Weather forecasting is such a system and hence one can’t forecast accurately for more than a few days irrespective of how much data you may gather and feed into a supercomputer. This phenomenon is more commonly referred to as the ‘butterfly effect’, where a metaphorical flutter of a butterfly's wings in the Amazon rainforests could cause a cyclone in the Bay of Bengal. Anyone who claims they can connect the flutter of the butterfly to the cyclone, and can actually predict it, is talking rubbish.

An AI vendor once confidently bragged to us that their tool could look at a job description, evaluate 100 CVs, and pick the best 5 suited for the job. When we probed a little deeper into how their algorithm worked, they resorted to the standard shenanigan of “we use a deep learning algorithm.” You probe a little deeper into what deep learning really means and they are often left gasping. When we got this tool to actually look at 100 odd CVs and shortlist the 5 best, we found a zero match with what a good recruiter and hiring manager shortlisted!

Professor Arvind Narayanan of Princeton has written that “Much of what’s being sold as ‘AI’ today is snake oil — it does not and cannot work.”
Why is this happening? How can we recognize flawed AI claims and push back?

He has classified AI into three broad buckets:

1. Areas where AI is genuine and making rapid progress – face recognition, medical diagnosis from scans, speech-to-text, reverse image search, etc.

2. Areas that are imperfect but improving – detection of spam, hate speech, copyright violation, etc.

3. Fundamentally dubious areas – predicting job success, recidivism, at-risk kids, etc.

The last category, which is really about predicting social outcomes, is essentially the snake oil being sold to gullible users and used as a pretext for collecting large amounts of data. Users are made to believe that magical insights can somehow be extracted from large amounts of data—and the more the data, the better the insights!

Professor Narayanan claims that there has been no real improvement in the third category, despite how much data you throw at it. He further goes on to show that for predicting social outcomes, AI is worse off than manual scoring using just a few features. I have no doubt that the inherent limitation in this area is imposed by chaos theory.

In Conclusion

‘Data rich’ is a phrase that has been misused of late to create false expectations. The protagonists of predicting social outcomes will no doubt claim that it is only a matter of time before it gets better. This is untrue. Some things can get better with time, but some ideas have inherent limitations.

If small differences in initial conditions, such as those due to rounding errors in numerical computation, can yield widely diverging outcomes even for deterministic systems—where an approximate present cannot determine an approximate future—imagine how much more indeterminate or irrelevant the predictions would be for inherently non-deterministic systems like social behaviors and outcomes!

Just as Heisenberg’s uncertainty principle places fundamental limitations at an atomic level, chaos theory places a similar limitation in areas like social outcomes.

Why this is finance’s second big transformation opportunity — and how to act without falling for the hype.

Let’s rewind to the early 2000s. Remember the ERP wave? SAP implementations that took months, even years. Cross-functional teams. Painful integration. Cultural overhaul.

And who was at the heart of it?
Finance!

CFOs led those massive ERP transformations. Not because we were tech experts, but because financial systems cut across the entire organization. We understood the workflows, the pain points, and ultimately — the decisions.

Today, AI is presenting us with a similar inflection point.
Only this time, it’s moving faster.
And once again, CFOs must take the lead.

ERP Was Yesterday. AI Is Today.

ERP touched everything — procurement, sales, operations, finance, HR. It standardized, streamlined, and created single sources of truth.

AI now offers something just as powerful: not standardization, but intelligence.

• Forecasts are no longer backward-looking.

• Reconciliations aren’t just automated — they’re self-correcting.

• Narrative reports can be drafted with a click.

• Cost optimization, risk assessment, ESG tracking — all can be AI-enhanced.

But it won’t happen on its own.
Just like ERP needed a cross-functional blueprint with finance at the center, AI needs a strategy — and a sponsor.

CFOs: Here’s What Leadership Looks Like This Time

1. Avoid the FOMO Trap

AI is evolving at breakneck speed. Updates drop every week.
Some companies ban everything except Microsoft Copilot. Others go wild with dozens of tools.

The smart CFO doesn’t chase hype. They ask:

“What real value can AI drive for us this year, next year, and in three years?”

Without that lens, AI becomes a distraction.
With it, it becomes a multiplier.

2. Define Your AI Strategy — Not a Tech Wishlist

This isn’t about collecting shiny tools. It’s about designing a value-first roadmap:

• Where can AI improve margins or profitability?

• Where can it reduce reliance on repetitive manpower?

• Where does AI make decisions faster or more accurate?

• What part of finance (FP&A, audit, controls, reporting) should pilot first?

Start there. Test fast. Scale what works.

3. Data Discipline = AI Power

Here’s the piece most people miss: AI models are only as smart as the data you feed them.

That means:

• Clean, tagged, structured data at source

• A unified data architecture across systems

• Understanding that training AI on your organizational data is what creates relevance and accuracy

You can’t expect powerful outputs if the system is learning from incomplete or inconsistent data.
Just like ERP needed standardization of formats, AI needs quality training data — and the finance function has to own this rigor.

Pilot, Then Scale

Not everything will work. That’s okay.
Your goal isn’t perfection. It’s to create internal proof points.

• Run a pilot in FP&A.

• Try AI for invoice matching.

• Test a margin analysis bot.

Show one real win, and you’ll have momentum.

What AI Changes in the CFO’s World

• Decisions accelerate — you’re not waiting for month-end.

• People are redeployed, not just reduced.

• Strategy gets sharper, because you finally have real-time, scenario-based views.

• Compliance and controls become proactive, not reactive.

But only if you lead the change.

Bottom Line

AI may feel overwhelming — but remember, you’ve done this before.

You led the charge during ERP. You understood the cross-org implications. You asked the hard questions. You brought the discipline.

This is that moment again.

CFOs who embrace AI — without the FOMO, with a strong strategic lens, and with control over data quality — will future-proof their organizations and redefine their own role in the C-suite.

Coming Up Next

How to Build Your AI Learning Curve as a CFO — Without Drowning in Jargon or Vendors

Ever stop to think that the code behind your favorite apps might be the secret sauce to their success? I'm talking about algorithms—the unsung heroes that decide what you binge-watch on Netflix, which job recommendations pop up on LinkedIn, and why TikTok seems to know exactly when you need a cat video break.

The Unlikely Origins of Our Digital Superpowers

Let’s kick this off with a little time travel. Back in 9th-century Baghdad, Muhammad al-Khwarizmi (yup, that’s where “algorithm” gets its name) was busy laying down the math rules that would eventually shape our data-driven world. Fast-forward to 1843, when Ada Lovelace penned what many consider the first algorithm for a machine—imagine dreaming up computers before they were even a twinkle in anyone’s eye! And of course, there’s Alan Turing, who, during WWII, cracked the Enigma code and turned the tide of history with his algorithmic genius. These pioneers set the stage for everything we now take for granted.

Why Algorithms Are Game-Changers

Today, algorithms are our digital lab assistants, working tirelessly behind the scenes to make our lives more efficient and our products smarter:

• Speed & Efficiency: They process gargantuan datasets in seconds, automating the mundane so you can focus on the creative, high-impact decisions.

• Scalability & Personalization: Look at Google’s PageRank, Netflix’s binge-worthy recommendations, and Amazon’s “Customers also bought…” feature. These algorithms don’t just scale operations—they tailor experiences, making every user feel like a VIP.

• Data-Driven Insights: For us product managers, algorithms are gold mines of actionable insights, revealing hidden patterns that drive innovation and streamline product development.

The Flip Side: Challenges to Keep in Check

But hey, it’s not all smooth sailing. Relying too much on algorithmic insights without a healthy dose of human intuition can sometimes lead companies down the wrong path. Here’s the lowdown:

• Historical Baggage: Algorithms learn from past data. If that data is biased or outdated, you might end up with features that are stuck in the past—goodbye, innovation!

• Opacity & Black Box Syndrome: Many algorithms are like mysterious wizards—they work their magic behind closed doors. This lack of transparency can be a real buzzkill, especially when it comes to making ethical decisions.

• Echo Chambers: When algorithms optimize based solely on user interactions, they can narrow your product’s focus. That’s great for engagement metrics, but it might mean missing out on broader opportunities.

• Over-Reliance: Data is powerful, but it can’t tell the whole story. Without qualitative insights and direct user feedback, you risk building products that miss the mark.

Balancing the Equation: Product Management in the Algorithm Age

So, what’s the secret sauce for product managers? It’s all about balance. Use algorithms to supercharge your insights, but never forget the human element. Here’s how:

1. Combine Data with Gut Feel: Use quantitative insights to inform your decisions, but always pair them with qualitative feedback, market trends, and a touch of creative intuition.

2. Stay Agile & Iterative: Treat algorithms as your testing ground. Use them to quickly prototype, iterate, and validate ideas before committing big resources.

3. Demand Transparency: Make sure the tools you use are regularly audited for bias and inaccuracies. Transparency isn’t just ethical—it builds trust.

4. Educate & Empower Your Team: Cultivate a culture of “data thinking” where everyone understands both the power and the pitfalls of algorithmic insights. Empower your team to question, validate, and iterate on what the data tells you.

Looking Ahead: AI, Your New Co-Pilot

Here’s where it gets really exciting. With AI turbocharging our algorithms, we’re no longer talking about rigid “if-else” chains. AI is making our systems dynamic, adaptable, and even creative. Imagine an AI co-pilot that predicts market shifts before they happen or fine-tunes your product features in real time based on user behavior. It’s not about replacing us—it’s about making us even better at what we do.

Final Thoughts

Algorithms are transforming our digital landscape, driving efficiency, personalization, and scalability like never before. But as product managers, our job is to harness this power without losing sight of the human touch. When we balance data with intuition, we not only build products that scale—we build products that truly resonate.