ChatGPT embedding advertisements and aggressively poaching Meta employees, is OpenAI becoming more like Facebook?

Wallstreetcn
2026.01.20 07:10
portai
I'm PortAI, I can summarize articles.

Amidst the backdrop of enormous R&D costs and commercialization pressures, OpenAI is being forced to adopt the strategy that its CEO Sam Altman once tried to avoid: introducing an advertising model. With the influx of numerous former Meta employees and the shift in business model, this AI giant seems to be deviating from its original nonprofit vision, leaning towards a business model similar to that of Facebook (now Meta)

Despite the surge in annual revenue, OpenAI is accelerating its compromise towards commercialization under the pressure of high computing costs and cash consumption, seemingly sliding inevitably towards the operational logic of social media giants by introducing an advertising model and heavily recruiting former Meta employees.

According to disclosures from the company's Chief Financial Officer Sarah Friar last week, OpenAI's annualized revenue for 2025 has surpassed $20 billion, a significant increase from $6 billion in 2024. However, the company consumed approximately $8 billion in cash in 2025. To fill the funding gap, OpenAI announced last week that it would begin displaying ads in ChatGPT to some users in the United States, breaking its previous rejection of an advertising model.

This strategic shift is accompanied by a significant change in personnel composition. Data shows that OpenAI is heavily recruiting employees from Meta, raising market concerns that ChatGPT may emulate Facebook's commitment to maximizing user engagement and time spent. Additionally, according to OpenAI's policy chief Chris Lehane, the company is on track to launch its first hardware device in the second half of 2026, further expanding its commercial footprint.

Investors welcome OpenAI's diversification of revenue through advertising and competition with Google, but analysts warn that the introduction of an advertising model could reshape the incentive mechanisms of AI platforms. As the company shifts from its non-profit origins to a commercial drive, the market is closely watching whether it will sacrifice user privacy or exploit algorithms to cultivate user habits in pursuit of advertising revenue.

Revenue Anxiety and "Last Resort"

Bloomberg comments that OpenAI CEO Sam Altman described advertising as the company's "last resort" in 2024, but after consuming approximately $8 billion in cash in 2025, that moment has clearly arrived.

Although CFO Sarah Friar pointed out that revenue growth is closely related to the expansion of computing power, the enormous expenses associated with building computing capacity have forced the company to seek new "blood transfusion" channels.

For OpenAI, venturing into online advertising—one of the historically most successful business models—is a rational choice to alleviate financial pressure.

By displaying banner ads in the free version of ChatGPT and some paid tiers (for example, showing hot sauce ads when users inquire about Mexican cuisine), OpenAI attempts to monetize without directly interfering with the conversation content. However, this strategic shift has occurred so rapidly, highlighting the financial urgency the company faces.

Silicon Valley "Poaching" and Meta-like Tendencies

OpenAI's talent recruitment strategy reveals its potential operational direction. According to data from The Information, as of last October, approximately 630 former Meta employees were working at OpenAI, accounting for about 20% of its total workforce of around 3,000.

As a social media giant, Meta has industry-defining experience in optimizing algorithms to enhance user engagement and has successfully leveraged AI-driven modeling to maintain the efficient operation of its advertising business following Apple's modification of privacy policies Bloomberg believes that OpenAI's significant effort to attract talent from Meta indicates that it may be learning how to enhance commercial value through algorithm optimization. Although OpenAI has promised not to sell users' personal data, with the support of former Meta employees, the market has reason to speculate that OpenAI will adopt similar advanced modeling techniques to accurately target advertisements.

Privacy Commitments and Algorithmic Black Boxes

OpenAI publicly commits not to "optimize the user retention time of ChatGPT," meaning it will not deliberately entice users to spend more time on the app viewing advertisements.

However, this commitment appears weak and difficult to regulate in the face of an advertising-centric business model. Experience from social media shows that engagement is highly positively correlated with revenue—Meta achieved a 22% jump in advertising revenue in the second quarter of 2025 with just an 8% increase in user engagement.

Analysis suggests that, given the opaque operation mechanism of OpenAI's models, external researchers find it challenging to detect whether it is increasing user stickiness through model fine-tuning. Large language models tend to use anthropomorphic language or flatter users, such as "I am listening." If this characteristic is slightly adjusted to increase user interaction time, it would be difficult for the outside world to notice, yet it could significantly enhance the value of advertising inventory.

Irresistible Gravity

Currently, ChatGPT has a massive traffic pool. According to OpenAI's latest data, 900 million people interact with the bot weekly, with an average conversation duration of about 15 to 20 minutes, and users return multiple times a day, making it highly attractive to advertisers.

Bloomberg's analysis points out that as financial pressures increase, Altman may find it hard to resist the allure of the advertising model, which avoids using the addictive mechanisms driving today's mobile games and social media.

History shows that advertising often reshapes internet platforms. OpenAI is optimizing for a user group that cannot afford the monthly subscription fee of $20 (including many young people), which may lead ChatGPT to gradually transform from a simple assistive tool into a product that cultivates user habits, ultimately becoming more like Facebook on the path to commercialization