AI security researchers from OpenAI, Anthropic, and different organizations are talking out publicly towards the “reckless” and “utterly irresponsible” security tradition at xAI, the billion-dollar AI startup owned by Elon Musk.
The criticisms observe weeks of scandals at xAI which have overshadowed the corporate’s technological advances.
Final week, the corporate’s AI chatbot, Grok, spouted antisemitic feedback and repeatedly known as itself “MechaHitler.” Shortly after xAI took its chatbot offline to handle the issue, it launched an more and more succesful frontier AI mannequin, Grok 4, which TechCrunch and others discovered to seek the advice of Elon Musk’s private politics for assist answering hot-button points. Within the newest growth, xAI launched AI companions that take the type of a hyper-sexualized anime woman and an excessively aggressive panda.
Pleasant joshing amongst staff of competing AI labs is pretty regular, however these researchers appear to be calling for elevated consideration to xAI’s security practices, which they declare to be at odds with trade norms.
“I didn’t need to put up on Grok security since I work at a competitor, however it’s not about competitors,” stated Boaz Barak, a pc science professor at the moment on depart from Harvard to work on security analysis at OpenAI, in a Tuesday put up on X. “I respect the scientists and engineers at xAI however the way in which security was dealt with is totally irresponsible.”
I did not need to put up on Grok security since I work at a competitor, however it’s not about competitors.
I respect the scientists and engineers at @xai however the way in which security was dealt with is totally irresponsible. Thread beneath.
— Boaz Barak (@boazbaraktcs) July 15, 2025
Barak significantly takes problem with xAI’s determination to not publish system playing cards — trade customary reviews that element coaching strategies and security evaluations in a superb religion effort to share data with the analysis group. Consequently, Barak says it’s unclear what security coaching was completed on Grok 4.
OpenAI and Google have a spotty popularity themselves in terms of promptly sharing system playing cards when unveiling new AI fashions. OpenAI determined to not publish a system card for GPT-4.1, claiming it was not a frontier mannequin. In the meantime, Google waited months after unveiling Gemini 2.5 Professional to publish a security report. Nonetheless, these firms traditionally publish security reviews for all frontier AI fashions earlier than they enter full manufacturing.
Techcrunch occasion
San Francisco
|
October 27-29, 2025
Barak additionally notes that Grok’s AI companions “take the worst points we at the moment have for emotional dependencies and tries to amplify them.” Lately, we’ve seen numerous tales of unstable individuals growing regarding relationship with chatbots, and the way AI’s over-agreeable solutions can tip them over the sting of sanity.
Samuel Marks, an AI security researcher with Anthropic, additionally took problem with xAI’s determination to not publish a security report, calling the transfer “reckless.”
“Anthropic, OpenAI, and Google’s launch practices have points,” Marks wrote in a put up on X. “However they no less than do one thing, something to evaluate security pre-deployment and doc findings. xAI doesn’t.”
xAI launched Grok 4 with none documentation of their security testing. That is reckless and breaks with trade greatest practices adopted by different main AI labs.
If xAI goes to be a frontier AI developer, they need to act like one. 🧵
— Samuel Marks (@saprmarks) July 13, 2025
The fact is that we don’t actually know what xAI did to check Grok 4. In a broadly shared put up within the on-line discussion board LessWrong, one nameless researcher claims that Grok 4 has no significant security guardrails based mostly on their testing.
Whether or not that’s true or not, the world appears to be discovering out about Grok’s shortcomings in actual time. A number of of xAI’s issues of safety have since gone viral, and the corporate claims to have addressed them with tweaks to Grok’s system immediate.
OpenAI, Anthropic, and xAI didn’t reply to TechCrunch request for remark.
Dan Hendrycks, a security adviser for xAI and director of the Heart for AI Security, posted on X that the corporate did “harmful functionality evaluations” on Grok 4, indicating that the corporate did some pre-deployment testing for security considerations. Nonetheless, the outcomes to these evaluations haven’t been publicly shared.
“It considerations me when customary security practices aren’t upheld throughout the AI trade, like publishing the outcomes of harmful functionality evaluations,” stated Steven Adler, an impartial AI researcher who beforehand led harmful functionality evaluations at OpenAI, in an announcement to TechCrunch. “Governments and the general public need to know the way AI firms are dealing with the dangers of the very highly effective methods they are saying they’re constructing.”
What’s fascinating about xAI’s questionable security practices is that Musk has lengthy been one of many AI security trade’s most notable advocates. The billionaire proprietor of xAI, Tesla, and SpaceX has warned many occasions in regards to the potential for superior AI methods to trigger catastrophic outcomes for people, and he’s praised an open strategy to growing AI fashions.
And but, AI researchers at competing labs declare xAI is veering from trade norms round safely releasing AI fashions. In doing so, Musk’s startup could also be inadvertently making a robust case for state and federal lawmakers to set guidelines round publishing AI security reviews.
There are a number of makes an attempt on the state degree to take action. California state Sen. Scott Wiener is pushing a invoice that may require main AI labs — possible together with xAI — to publish security reviews, whereas New York Gov. Kathy Hochul is at the moment contemplating an identical invoice. Advocates of those payments notice that almost all AI labs publish this sort of data anyway — however evidently, not all of them do it constantly.
AI fashions right this moment have but to exhibit real-world eventualities during which they create actually catastrophic harms, such because the demise of individuals or billions of {dollars} in damages. Nonetheless, many AI researchers say that this may very well be an issue within the close to future given the fast progress of AI fashions, and the billions of {dollars} Silicon Valley is investing to additional enhance AI.
However even for skeptics of such catastrophic eventualities, there’s a robust case to counsel that Grok’s misbehavior makes the merchandise it powers right this moment considerably worse.
Grok unfold antisemitism across the X platform this week, only a few weeks after the chatbot repeatedly introduced up “white genocide” in conversations with customers. Quickly, Musk has indicated that Grok shall be extra ingrained in Tesla automobiles, and xAI is making an attempt to promote its AI fashions to The Pentagon and different enterprises. It’s exhausting to think about that folks driving Musk’s vehicles, federal employees defending the U.S., or enterprise staff automating duties shall be any extra receptive to those misbehaviors than customers on X.
A number of researchers argue that AI security and alignment testing not solely ensures that the worst outcomes don’t occur, however additionally they shield towards near-term behavioral points.
On the very least, Grok’s incidents are inclined to overshadow xAI’s fast progress in growing frontier AI fashions that greatest OpenAI and Google’s know-how, only a couple years after the startup was based.
