Blab News
Chamath Palihapitiya Criticizes Metas AI, Rejects Job-Loss Alarm, Admits SPAC Incentive Misalignment
On a recent episode of The Axios Show, venture capitalist Chamath Palihapitiya—known for his role in building Facebook and for leading the investment firm Social Capital—cut a sharp critique of Meta Platforms Inc. and offered a candid look at his own business practices. He argued that Meta had squandered a prime opportunity to dominate artificial intelligence, downplayed fears of an AI‑driven job apocalypse, and confessed that the incentive structure in his special‑purpose acquisition (SPAC) deals had been fundamentally misaligned.
Palihapitiya’s focus was Meta’s Llama family of large‑language models. He said the company possessed a “huge chance to lead in AI” because of its vast user data and worldwide distribution network. Yet, when the wave of chatbot products burst onto the scene in 2023, Meta, according to Palihapitiya, missed the moment and let competitors such as Nvidia’s Jensen Huang and open‑weight Chinese laboratories seize the market. He outlined the AI landscape as split into three pillars: closed American labs, open‑weight Chinese challengers, and an open‑source American lane that Meta should have dominated. Palihapitiya noted that while Meta’s Llama models are open‑weight, they are not fully open source, and the company has shifted toward proprietary offerings.
The comments arrive after Meta announced in February 2024 that it had spent $35.3 billion on research and development in 2023, making it the third‑largest R&D spender worldwide. The company has released several Llama versions, including Llama 4 in April 2025, and has built an AI assistant called Meta AI on top of the models.
When asked about the possibility that AI could eliminate large numbers of jobs, Palihapitiya dismissed the idea as an “incredible headline.” He argued that the argument ignores historical patterns, where new technology has created more work than it has destroyed. While acknowledging that some roles—such as customer support—may disappear, he said the overall effect is a shift in how people allocate their time. The investor referenced statements from OpenAI’s Sam Altman, who has also said a full‑scale job apocalypse is unlikely.
The most surprising part of the interview was Palihapitiya’s admission that the incentives he built into his SPAC deals were misaligned. He said he had resisted acknowledging the problem for years because he was “too insecure to admit it.” He explained that his compensation structure allowed him to earn money regardless of the long‑term performance of the companies he merged with. He added that while he does not believe he made bad deals, the stock performance of those companies was a clear indicator of the value created.
To address the issue, Palihapitiya announced a new SPAC vehicle, American Exceptionalism Acquisition Corp. In that structure, his shares vest only if the merged company trades at least 50 % above its listing price, roughly $15 per share. The change ties his payout to investor returns, a link that was missing in his earlier deals. He hinted that a target company may be in sight, though due diligence is still pending. The timing follows a resurgence of SPAC activity, with 251 blank‑cheque firms seeking targets and nearly $47 billion in capital.
Palihapitiya also touched on privacy, suggesting that AI companies should sometimes know who a user is and what they are prompting. He defended President Trump’s immigration policies, noting his own immigrant background. He is also the founder of an AI startup called 8090, indicating that his views are informed by active involvement in the technology sector.
In summary, Palihapitiya’s interview highlights three key points: Meta’s perceived failure to seize an AI leadership position, a skeptical view of an AI‑driven job apocalypse, and a candid admission that his SPAC incentive structure was misaligned. He has taken steps to redesign his SPAC model to better align with investor outcomes. The broader tech community will watch how Meta responds to the criticism and whether it can regain a leading role in the evolving AI landscape.
Palihapitiya’s focus was Meta’s Llama family of large‑language models. He said the company possessed a “huge chance to lead in AI” because of its vast user data and worldwide distribution network. Yet, when the wave of chatbot products burst onto the scene in 2023, Meta, according to Palihapitiya, missed the moment and let competitors such as Nvidia’s Jensen Huang and open‑weight Chinese laboratories seize the market. He outlined the AI landscape as split into three pillars: closed American labs, open‑weight Chinese challengers, and an open‑source American lane that Meta should have dominated. Palihapitiya noted that while Meta’s Llama models are open‑weight, they are not fully open source, and the company has shifted toward proprietary offerings.
The comments arrive after Meta announced in February 2024 that it had spent $35.3 billion on research and development in 2023, making it the third‑largest R&D spender worldwide. The company has released several Llama versions, including Llama 4 in April 2025, and has built an AI assistant called Meta AI on top of the models.
When asked about the possibility that AI could eliminate large numbers of jobs, Palihapitiya dismissed the idea as an “incredible headline.” He argued that the argument ignores historical patterns, where new technology has created more work than it has destroyed. While acknowledging that some roles—such as customer support—may disappear, he said the overall effect is a shift in how people allocate their time. The investor referenced statements from OpenAI’s Sam Altman, who has also said a full‑scale job apocalypse is unlikely.
The most surprising part of the interview was Palihapitiya’s admission that the incentives he built into his SPAC deals were misaligned. He said he had resisted acknowledging the problem for years because he was “too insecure to admit it.” He explained that his compensation structure allowed him to earn money regardless of the long‑term performance of the companies he merged with. He added that while he does not believe he made bad deals, the stock performance of those companies was a clear indicator of the value created.
To address the issue, Palihapitiya announced a new SPAC vehicle, American Exceptionalism Acquisition Corp. In that structure, his shares vest only if the merged company trades at least 50 % above its listing price, roughly $15 per share. The change ties his payout to investor returns, a link that was missing in his earlier deals. He hinted that a target company may be in sight, though due diligence is still pending. The timing follows a resurgence of SPAC activity, with 251 blank‑cheque firms seeking targets and nearly $47 billion in capital.
Palihapitiya also touched on privacy, suggesting that AI companies should sometimes know who a user is and what they are prompting. He defended President Trump’s immigration policies, noting his own immigrant background. He is also the founder of an AI startup called 8090, indicating that his views are informed by active involvement in the technology sector.
In summary, Palihapitiya’s interview highlights three key points: Meta’s perceived failure to seize an AI leadership position, a skeptical view of an AI‑driven job apocalypse, and a candid admission that his SPAC incentive structure was misaligned. He has taken steps to redesign his SPAC model to better align with investor outcomes. The broader tech community will watch how Meta responds to the criticism and whether it can regain a leading role in the evolving AI landscape.
