The race for "AI Supremacy" is over — at least for now.
We recorded 17 claims from this post. 4 were checked against evidence. 13 can’t be checked. 24 more are held. Read the post at its source.
Claims from this post (17)
In the order they appear in the post. Each quotation is the speaker’s words in this post. A label such as “Supported” or “Can’t be checked” is the claim’s current assessment; its page has the evidence and the reasons.
Assessment: UnresolvedSaid
Observation
“In late December, a company called DeepSeek, apparently initially built for quantitative trading rather than LLMs, produced a nearly state-of-the-art model that required only roughly 1/50th of the training costs of previous models, — instantly putting them in the big leagues with American companies like OpenAI, Google, and Anthropic, both in terms of performance and innovation.”
Our reading: In late December, DeepSeek, apparently first built for quantitative trading, produced a nearly state-of-the-art model that needed roughly 1/50th of the training costs of previous models, putting it in the big leagues with American companies such as OpenAI, Google and Anthropic in both performance and innovation.
No horizon Evidence as of
Limitation: Unknown accounting basis prevents a supported or contradicted financial verdict.
Assessment: SupportedSaid
Forecast
“None of this means however that China won the AI race or even took the lead. American companies will incorporate these new results and continue to produce new results of their own.”
Our reading: The recent Chinese results do not mean China has won the AI race or even taken the lead; American companies will incorporate these results and keep producing new results of their own.
Settled before its deadline Evidence as of
Limitation: The decisive examples involve model checkpoints and generated training data, rather than proof that every named technique was adopted.
Assessment: Can’t be checkedSaid
Causal claim
“Others may catch up, too, because LLMs just got a lot cheaper, and consequently the requirement for vast arrays of special purpose hardware has somewhat diminished.”
Our reading: Because LLMs got much cheaper and need less special hardware, other players may catch up.
No horizon
Assessment: UnresolvedSaid
Observation
“Meanwhile, revenue (order of $5B) has been modest, compared to outlays that have been much greater; no actual killer app (relative to the costs) has emerged.”
Our reading: OpenAI's revenue (on the order of $5B) has been modest compared with much greater outlays, and no actual killer app, relative to the costs, has emerged.
No horizon Evidence as of
Limitation: Contemporary annual figures were projections [e1, e2].
Assessment: Can’t be checkedSaid
Forecast
“I have said it before, and I will say it again—OpenAI may well become the WeWork of AI.”
Our reading: OpenAI may well become the WeWork of AI.
No horizon
Assessment: Can’t be checkedSaid
Forecast
“Nvidia could soon take a serious hit, too, for two reasons. First, the DeepSeek results suggest that large language models might be trained far more efficiently going forward.”
Our reading: Nvidia could soon take a serious hit because LLMs may be trainable far more efficiently.
No horizon
Qualifier: LLMs can be trained far more efficiently
Assessment: Can’t be checkedSaid
Recommendation or value judgment
“Building $500B worth of power and data centers in the service of enormous collections of those chips isn't looking so sensible, either.”
Our reading: Building $500B worth of power and data centers to serve enormous collections of top Nvidia chips isn't looking so sensible either.
No horizon
Assessment: Can’t be checkedSaid
Conditional forecast
“If all this is true, the CHIPS Act, which was designed to slow China in the AI race, may turn out to be one of the worst backfires in history.”
Our reading: If all this is true, the CHIPS Act, which was designed to slow China in the AI race, may turn out to be one of the worst backfires in history.
No horizon
Only if: if all this is true
Assessment: Can’t be checkedSaid
Recommendation or value judgment
“(Trump’s boosting of Stargate seemingly fits with the same magical thinking in LLMs, premised on the same hope of achieving a supremacy that may never come).”
Our reading: Trump's boosting of Stargate seemingly fits with the same magical thinking about LLMs, premised on the hope of achieving a supremacy that may never come.
No horizon
Qualifier: seemingly
Assessment: UnresolvedSaid
Causal claim
“We accidentally upped their technical game.”
Our reading: US export controls accidentally improved China's AI technical capabilities by forcing efficiency innovations.
No horizon Evidence as of
Limitation: Technical development accounts come primarily from DeepSeek.
Assessment: Can’t be checkedSaid
Causal claim
“And maybe kneecapped our greatest silicon company, Nvidia.”
Our reading: US export controls may have kneecapped Nvidia.
No horizon
Assessment: Can’t be checkedSaid
Conditional forecast
“If the race to LLMs continues to be unregulated in the US, and LLMs remain detached from reality, rushed development cycles and intense global competition might well exacerbate risks of misinformation, biased outputs, privacy breaches, and misuse by malicious actors.”
Our reading: If the LLM race stays unregulated in the US and LLMs remain detached from reality, rushed development cycles and intense global competition might well worsen the risks of misinformation, biased outputs, privacy breaches and misuse by malicious actors.
No horizon
Only if: the race to LLMs continues to be unregulated in the US
Assessment: Can’t be checkedSaid
Recommendation or value judgment
“One that deserves Congressional investigation was Meta's decision to open source their LLMs.”
Our reading: Congress should investigate how Meta's open-sourcing of its LLMs helped China catch up.
No horizon
Assessment: Can’t be checkedSaid
Forecast
“Putting so many eggs in Altman’s basket, as the White House did last week and others have before, may also prove to be a mistake in hindsight.”
Our reading: The White House's heavy reliance on Altman may prove a mistake in hindsight.
No horizon
Assessment: Can’t be checkedSaid
Recommendation or value judgment
“Altman may be a master salesman, but Musk is correct that the US should not be so reliant on him and should not have given him such an eminent seal of approval based on so little.”
Our reading: Marcus agrees with Musk that the US should not be so reliant on Altman and should not have given him such an eminent seal of approval based on so little.
No horizon
Qualifier: based on so little
Assessment: Can’t be checkedSaid
Recommendation or value judgment
“Furthermore, as I have argued for years, LLMs are too opaque, too unwieldy, and too difficult to debug and verify. The answer lies elsewhere; betting our future on that single idea is foolish.”
Our reading: LLMs are too opaque, unwieldy and hard to debug and verify, so betting the future on them is foolish.
No horizon
Assessment: Can’t be checkedSaid
Forecast
“The race to AGI will be won not by the country with the most chips but by the one that best fosters true innovation.”
Our reading: The country that best fosters true innovation, not the one with the most chips, will win the race to AGI.
No horizon
Held from this post (24)
These records did not pass every check, so nothing about them is published. Held is not a finding.
- The record does not meet the current rules. Our reading, not assessed: OpenAI no longer clearly leads technically; competitors have largely caught up and often undercut it on price.
- The two reviews reached different assessments. Our reading, not assessed: LLMs are about to get a lot cheaper, possibly making consumers the big winners, despite reliability problems.
- The two reviews read the claim differently
- The record does not meet the current rules. Our reading, not assessed: The US and China are quickly converging on a tie in AI.
- The criteria are ambiguous. Our reading, not assessed: DeepSeek's results suggest competitive AI is possible without Nvidia's top-of-the-line chips.
- The two reviews reached different assessments. Our reading, not assessed: OpenAI's relationship with Microsoft has soured and Altman's credibility has diminished.
- The two reviews reached different assessments. Our reading, not assessed: Marcus bets that, other things being equal, in three years models will have continued to get more efficient and less expensive, but hallucinations and reliability problems will persist.
- The two reviews read the claim differently
- The two reviews reached different assessments. Our reading, not assessed: Marcus bets that, other things being equal and contrary to Silicon Valley scuttlebutt, neither the US nor China will achieve AGI by the end of 2027.
- The two reviews read the claim differently
- The two reviews reached different assessments. Our reading, not assessed: Marcus bets that, other things being equal, in three years AI advances will be more incremental than before and quickly matched.
- The two reviews read the claim differently
- The two reviews reached different assessments. Our reading, not assessed: There is almost no moat in LLMs; technical leads last months or weeks, not years.
- The two reviews read the claim differently
- The two reviews read the claim differently
- The two reviews read the claim differently
- The record does not meet the current rules. Our reading, not assessed: DeepSeek's r1 is a competitive, though not fully adequate, alternative to OpenAI's o1.
- The two reviews read the claim differently
- The two reviews read the claim differently
- The two reviews read the claim differently
- The record does not meet the current rules. Our reading, not assessed: Recent Chinese AI papers showed that the US is not decisively ahead in AI.
- The two reviews read the claim differently
- The two reviews reached different assessments. Our reading, not assessed: Marcus bets that, other things being equal, GPT-5 or a similarly impressive model will come eventually, perhaps from OpenAI, a Chinese company or a competitor like Google, and whoever gets there first will have only a short-lived advantage.
- The two reviews reached different assessments. Our reading, not assessed: GPT-5 has not been released as of January 2025.