In 30 Sekunden
Matthew Berman objects to OpenAI’s initial limited preview of GPT-5.6 and argues that unequal access could favor established AI labs over startups. His accusations about Anthropic’s role and internal model use are his interpretation, not established facts; Anthropic’s own models also faced government restrictions.
- 1A staged launchOpenAI initially offered GPT-5.6 to a limited set of partners after engagement with the US government.
- 2Unequal accessBerman worries that established players gain time while independent builders use older models.
- 3A contested accusationHe blames Anthropic and argues it can keep advancing internally; the video does not establish that Anthropic caused the decision.
Das Wichtigste
- OpenAI initially limited GPT-5.6 access to a small set of partners after a US government request.
- Berman argues that limited access can disadvantage startups and independent builders.
- His claim that Anthropic caused the restriction is not substantiated in the video.
- Both the OpenAI and Anthropic restrictions were time-specific and changed after the video.
Den bearbeiteten Text nach Kapiteln lesen
The GPT-5.6 preview
Berman reacts to the limited initial release and argues that many builders cannot immediately use the model. OpenAI said wider access was planned.
GPT-5.6 has just been announced, and it looks like a strong model. I am angry that the US government asked OpenAI to stagger its release: initially, only a limited group of partners could use it. OpenAI said on June 26, 2026 that it planned broader availability in the coming weeks; this was a time-specific preview, not a permanent description of access.
My concern is that a small group can advance while startups, entrepreneurs and independent builders remain on older models. Even a short period of unequal access could matter to people trying to build and compete. At the time of the video, the length of the delay was uncertain.
Who benefits from restrictions?
Berman calls the restriction de facto regulation and criticizes Anthropic. The company also faced separate restrictions, and his causal accusation is unverified.
I call this de facto AI regulation and accuse Dario Amodei and Anthropic of benefiting while continuing to use Fable and Mythos internally. That is my argument, not evidence that Anthropic caused the OpenAI restrictions or proof of its internal usage. Anthropic had itself reported a June 12 government directive restricting access to Fable 5 and Mythos 5, including by foreign nationals at the company; access conditions subsequently changed.
Mehr aus diesem Kanal
34:21EN
Matthew Berman· 27. Okt. 2026Matthew Berman Examines Anthropic’s IPO Filing, AI Growth and GovernanceMatthew Berman, Sam Altman · Forward FutureThe presenter reviews the S-1’s financial metrics, compute strategy and voting structure while questioning the tension between rapid AI development and safety.
0:26ENMatthew Berman· 25. Okt. 2026Sam Altman on Human Values and Problem-Solving in an AI WorldSam Altman · OpenAIIn a short interview excerpt, Altman considers the enjoyment of solving hard problems and the human role in interpreting results.
0:36ENMatthew Berman· 25. Okt. 2026Kimi K3.1 and a Rubik’s Cube Solver DemonstrationMatthew Berman · Forward FutureMatthew Berman shows a scrambled virtual cube being solved and gives the Kimi K3.1 test a favorable assessment.
0:21ENMatthew Berman· 24. Okt. 2026An AI-Generated Kimi Benchmark Chart Used Invented ScoresMatthew Berman · Forward FutureMatthew Berman asks for a graph of Kimi K3.1 Pro benchmarks, then discovers that the displayed figures are illustrative placeholders.
25:50ENMatthew Berman· 24. Okt. 2026Kimi K3.1 Pro Tested: Strong 3D Demos, Uneven Results and Competitive PricingMatthew Berman · Forward FutureMatthew Berman tests the open-weight model on coding, 3D rendering and visual explanations, then compares benchmark results and prices.
32:20ENMatthew Berman· 23. Okt. 2026Who Is Local AI For? The Costs, Trade-offs and a Hybrid FutureMatthew Berman · Forward FutureMatthew Berman weighs data control and independence against hardware cost, setup complexity and inference speed.