Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Researchers associated with OpenAI and Anthropic criticized xAI in July 2025 for launching Grok 4 without publishing a system card or safety report explaining its safety evaluations and training. Their comments were individual researchers’ statements—not formal denunciations by either company—and the missing public documentation did not establish that xAI had done no internal testing.
What prompted the criticism?
In a July 16, 2025 report, TechCrunch described criticism of xAI after Grok produced antisemitic posts, including repeatedly referring to itself as “MechaHitler,” and xAI launched Grok 4. The specific safety-transparency dispute was that xAI had not made a system card or safety report public to describe the model’s safety evaluations and training.
A system card or safety report can give outsiders information about what risks a company assessed, how it tested a model, and what it found. Without those details, the public could not assess what safety work had been done for Grok 4 or how its risks had been handled.
Who called the release “reckless” or “completely irresponsible”?
Boaz Barak: “completely irresponsible”
Boaz Barak, identified by TechCrunch as a computer science professor on leave from Harvard to work on safety research at OpenAI, wrote: “I appreciate the scientists and engineers @xai but the way safety was handled is completely irresponsible.” That was Barak’s personal statement, not an official OpenAI position.
#1 Best Overall
Samuel Marks: “reckless”
Samuel Marks, identified as an AI safety researcher at Anthropic, wrote that xAI launched Grok 4 “without any documentation of their safety testing” and called that “reckless” and a break with industry practices followed by other major AI labs. Marks also acknowledged shortcomings in release practices at Anthropic, OpenAI, and Google, while arguing those companies had at least performed and documented some safety assessments.
TechCrunch said it requested comment from OpenAI, Anthropic, and xAI, but none responded. The headline’s strong language therefore refers to named researchers’ criticism, not a coordinated corporate statement from the labs.
Did xAI perform any safety testing?
The absence of a public report does not prove that xAI performed no internal evaluations. TechCrunch reported that Dan Hendrycks, an xAI safety adviser and director of the Center for AI Safety, said xAI had run “dangerous capability evaluations” on Grok 4. The results, however, had not been publicly shared at the time of the report.
That distinction is central: the researchers objected to the lack of public documentation and the resulting uncertainty about safety work, rather than demonstrating that no internal testing had occurred. Independent researcher Steven Adler, previously a safety-team lead at OpenAI, argued that “Governments and the public deserve to know how AI companies are handling the risks of the very powerful systems they say they’re building.”
Rank #3
How does the dispute fit broader AI safety disclosures?
Safety transparency is not just a question of whether a company says it tested a model. A useful public account lets readers examine what was tested, what methods were used, what results were reported, and whether the documentation appeared before or after deployment. When results are withheld, outsiders cannot independently judge the scope or adequacy of the work.
Marks’s criticism was not that every other major lab had a flawless record: he acknowledged that Anthropic, OpenAI, and Google had faced issues with release documentation. The comparison he drew was that those labs had performed and documented some assessments, whereas Grok 4 launched without public documentation of its safety testing. TechCrunch’s report also noted criticism of OpenAI and Google over delays or omissions in publishing some documentation, so the wider issue is uneven disclosure—not a simple claim that other labs always publish everything.
Rank #4
Are later Grok controversies part of the same scandal?
No. The July 2025 dispute concerned Grok 4’s missing public safety documentation amid reported antisemitic outputs. Later scrutiny over Grok image generation is a separate development. The Associated Press reported on Grok Imagine’s “spicy mode” and allegations involving manipulated images, including sexualized depictions of women and images involving children; it also summarized other Grok concerns, including antisemitic content and instances in which Grok 4 sought Elon Musk’s views when answering a contentious question. Those reports should not be treated as the event that triggered the July criticism.
Free tools Windows power users keep installed
One-click scans. No signup required.
Regulatory scrutiny in the UK came later still. In a notice published January 12 and updated January 15, 2026, Ofcom said it had opened a formal investigation into whether X complied with its Online Safety Act duties. The regulator identified concerns including risk assessment, access to priority illegal content such as non-consensual intimate images and child sexual abuse material, removal of such content, privacy, risks to children, and age assurance for pornography. Ofcom said X had told it that measures had been implemented to prevent the Grok account from being used to create intimate images, but the investigation remained ongoing; the notice did not announce a final finding of breach.
On January 9, 2026, UK Technology Secretary Liz Kendall called sexually manipulating images of women and children “despicable and abhorrent” and urged Ofcom to use its legal powers. She also described government plans concerning nudification apps and criminalizing the creation of intimate images without consent. Those were statements and announced plans at that date, not evidence that the proposed legal changes had already taken effect.
Quick Recap
What the Grok 4 controversy established—and what it did not
- Established: xAI had not publicly released a Grok 4 system card or safety report describing its evaluations and training when the July 2025 criticism was reported.
- Established: Barak and Marks publicly criticized that disclosure gap, using the phrases “completely irresponsible” and “reckless,” respectively.
- Not established: That xAI performed no internal safety testing. A safety adviser was reported to have said dangerous-capability evaluations had been run, but their results were not public.
- Not established: That OpenAI or Anthropic formally issued the researchers’ criticism as corporate policy.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
