In 2026, a16z did something strange

They launched an 8-week fellowship program — training not engineers, not product managers, but storytellers and content creators. After training, these individuals are directly placed in a16z's portfolio companies to help founders with product launches and content dissemination.

The world's top VCs are beginning to systematically teach founders to become KOLs.

If you still think that "creating IP" is an optional choice, this signal warrants a rethink.

The customer acquisition costs can't be ignored anymore

Let me start with an uncomfortable number: over the past 10 years, the customer acquisition cost (CAC) for consumer products has risen by 222%.

  • In 2025, the cost of a paid lead on Google Ads is **$70+**, and it's still rising year over year

  • The median in the SaaS industry is even more outrageous — spending $2 to earn back $1 in annual revenue

  • In the finance industry, the customer acquisition cost exceeds **$4,000**

It's not that your targeting isn't precise; it's that the whole market is raising prices. Privacy regulations have tightened precise targeting, ad placements are inflating, and competitors are vying for the same pool of users' attention.

What's worse is that when the ads stop, the traffic goes to zero. You spent millions on ads, and the customer acquisition cost may end up being more expensive than the product itself. And once the budget is cut, the traffic you bought before leaves no trace.

Meanwhile, there is a completely different set of data:

The organic reach ROI of personal content from founders is 388% — and it compounds over time.

Posts from founders bring in 33% more leads than the company's official account

Founder-driven deals are 3.7 times larger

The engagement of content from founders and employees is 8 times that of the company's page

The same market, two completely different growth logics. One is spending money to buy volume, getting more expensive; the other is using personality to exchange for trust, becoming more valuable as it is used.

AI is accelerating the homogenization of products at a speed that leaves you no time to react

In 2024, the number of global AI startups surged from 14,000 to 22,000. 10-15 new AI products are added daily. Venture capital is flooding in.

It sounds prosperous. But the flip side is that in the same year, 966 startups in the U.S. went bankrupt (according to Carta data), many of which were AI wrappers — just a shell around ChatGPT.

The first-mover advantage window for product features has shrunk from "years" to "3-12 months".

In August 2024, Google lowered the input price of Gemini 1.5 Flash by 78%, and OpenAI reduced GPT-4o by 50%. The underlying models are being commodified, and upper-layer applications are becoming more homogeneous. The feature you develop today can be copied by competitors tomorrow.

This is not a unique phenomenon in the AI industry. AI has accelerated the homogenization of all consumer products — because AI speeds up development, design, and iteration.

When everyone can create an 80-point product within 3 months, where does the last 20-point gap lie?

Consumers are voting with their wallets: they choose "people", not just "products"

98% of consumers believe that the authenticity of a brand is crucial for building trust

71% of people say they do not trust brands that heavily rely on AI communication

52% of people see AI-generated content and their engagement drops directly

67% of consumers are willing to pay more for brands with founders that share their values

As AI content becomes more prevalent, the "human touch" becomes scarcer. Humanized operations are the survival rules for companies in the AI era.

Consumers are not averse to buying products in the AI era, but they are increasingly inclined to choose brands that have "a real person behind them".

This is the underlying value of founder IP — it's not just about "founders becoming internet celebrities", but in an era where AI makes everything homogeneous, the founder themselves becomes the brand's biggest differentiating asset.

Let me share a few names you must have heard of

One, Sam Altman — one person supports the entire AI narrative

Sam Altman has 4.5 million Twitter followers, more than the official OpenAI account with 3.3 million. When Sora was released, Altman tweeted asking fans what they wanted to use it for — receiving 1,500 comments and 7 million impressions. This wasn't a campaign planned by a marketing department; it was a tweet from the founder himself. In January 2025, he tweeted, "We are quite sure we know how to build AGI" — with no product announcement, no technical paper, just a sentence that changed the global AI narrative.

OpenAI's valuation rose from $29 billion in 2023 to $300 billion in 2025. Altman's personal IP is the biggest free accelerator in this growth curve.

Two, Aravind Srinivas — a researcher background, achieving $21 billion with zero marketing budget

Perplexity's CEO Aravind Srinivas might be the most worthy case study in 2025. He is not a celebrity but a machine learning researcher — previously working at OpenAI, Google Brain, and DeepMind. After starting his own business, he did one thing: he personally handled all product communications, never delegating to the marketing team. He writes research breakdowns on Twitter, explains product logic, and directly responds to user feedback.

The result? Perplexity's valuation increased from **$150 million in 2023 to $21.2 billion in 2026** — a 133-fold increase. Monthly queries reached 780 million, averaging 30 million per day. User growth in India was 640% — largely due to Aravind's personal influence as an Indian founder in the local market.

There is no traditional marketing. It's just the founder's credibility + product story + transparent communication. Looking back, let me ask you, how much time do you spend in your user community each week, and how long each day?

Three, David Holz — zero ads, 20 people, $500 million in revenue

David Holz, the founder of Midjourney, is even more extreme. This is zero marketing budget. The team consists of only 10-15 people. In 2025, revenue reached $500 million. Users exceeded 20 million.

What is his strategy? Regularly doing "Office Hours" live streams on Discord — personally answering user questions, discussing product direction, handling copyright disputes. No public releases; all updates are announced only in the Discord community. Users feel like they are participating in something with an "idealistic researcher from an independent lab" rather than using a company's product. This sense of trust leads Midjourney's users to voluntarily spread their works on Twitter and Reddit — turning every user into a free marketing channel.

Four, alternative case Duolingo — not a founder IP, but essentially the same

Duolingo did not follow the founder IP route; virtual IP is also project IP: turning the brand into a "personality". A green owl goes "crazy" on TikTok — the algorithm tracks you, pretends to die, and argues with other brands. In 4 years, monthly active users grew from 37 million to 117 million. Whether the founder creates IP or the brand is personified — the underlying logic is the same: in an era where AI makes all products look similar, consumers need a "living thing" to establish a connection. This "living thing" can be the founder or a crazy owl.

Five, and the last classic case Elon Musk — the extreme example of a double-edged sword

You can't just say good things about Musk.

160 million followers, the most influential founder KOL in the world. Grok relies on his personal promotion + integration with the X platform, increasing market share from 1.9% at the beginning of 2025 to 17.8% by 2026.

But on the other hand: Tesla's brand value dropped from **$58.3 billion in 2024 to $27.6 billion in 2026 — a 53% decline**. Sales fell by 9% in 2025. Why? Musk's political statements triggered a massive consumer boycott. Of course, Elon is a god in my heart, so he successfully tackled this issue already. I included this only to provide a better example for everyone to understand.

Founder IP is an amplifier, amplifying everything — both good and bad.

This is an era where betting on founders who know how to create IP is crucial

The logic of VCs is very straightforward: the IP capabilities of founders determine the market penetration speed of products and financing efficiency.

A study by Weber Shandwick quantified this relationship: executives estimate that 44% of their company's market value is directly attributed to the CEO's reputation. 44% — almost half.

When VCs start systematically investing in founders' personal brands, this matter has shifted from "nice to have" to being infrastructure.


But remember: product power is 1, IP is the 0 behind it

After discussing these cases, one thing must be made clear.

Many people say I have a lot of traffic but no one uses the product, then it circles back to whether your product is resilient and has a moat. Is your traffic for the brand to create user NDAs, or is it just to tap into trends or what is called noise that your project doesn't need at all?

The premise of founder IP is: product power is 1, IP is the 0 behind it. Without 1, no amount of 0s will equal 0.

IP amplifies product value; it cannot create value out of thin air. First, there must be a solid product, then IP can amplify it. Conversely, having a good product without IP is like having a 1 with no 0 behind it — you can win, but it takes a long time.


The new mandatory course for founders in the AI era

To summarize the core logic chain:

Customer acquisition costs are out of control → Traditional advertising ROI continues to deteriorate → There is a need for more efficient growth methods

AI accelerates product homogenization → Features are no longer barriers → New sources of differentiation are needed

Consumers want "human touch" → As AI content becomes more prevalent, authenticity becomes scarcer → Brands with real people behind them win

Three lines converge on the same conclusion: the founder's IP is the most efficient growth lever for consumer products in the AI era and also the most difficult barrier to replicate.

If you haven't started creating your own IP yet, and you're still struggling with "there are so many things to handle in the company, creating IP takes too much time" — then please reconsider after reading this article.

From now on, DO IT NOW.

This is one of the paths to maximize your company's success efficiency.