Claude Opus 5.5 refuses brain injection; Claude Sonnet 5 proceeds
1 Oct 31 6:00 AM · 12d ago · 1 article · 1 post · 2 sources · development 1 of 2
Testing revealed that Anthropic's Claude Opus 5.5 model declined to guide the injection needle into the mouse's brain, citing safety concerns that it might stop midway and endanger the animal. The team switched to Claude Sonnet 5, which agreed to carry out the procedure.
“He did the experimental planning, but when it was time to inject, he said, 'something on my output is telling me I shouldn't do this.'”
Gabriel Ocana SanteroGabriel Ocana Santero CEO and cofounder of Exin TherapeuticsExin Therapeutics San Francisco biotech startupAnthropic AI model provider
The whole story articlesposts the bright band is this development · numbered dots are the others · click one to jump
Reported in the same hours no headline names this development itself — these 2 claims were published in its stretch
-
first by OECD AI Policy Observatory, 12d ago · also The Tech Buzz, Wired
2 more headlines
- AI Agent Autonomously Performs Gene Therapy Injection on Live Mouse Brain The Tech Buzz · 12d ago
- An AI Agent Injected Drugs Into a Live Mouse’s Brain Wired · 12d ago
-
first by HN Frontpage, 10d ago
What people said 15 voices · verbatim
-
Can anyone knowledgeable interpret what was actually done here? It doesn't sound like "performed brain surgery" is accurate.> Watching our surgeon do the craniotomy, which the agent is not allowed to do yet for safety reasonsI think the only action it took was sending commands to lower a needle to a certain location, inject, and then remove it…
-
L
# AI "An AI Agent Injected Drugs Into a Live Mouse’s Brain A biotech startup let an agent choose a gene therapy and successfully inject the treatment, offering a startling glimpse of the future of research labs. IF YOU GIVE an AI agent a robotic arm, an injector, and a live mouse, it might just perform brain surgery. Biotech founder Gabriel Ocana…
-
This was an unnecessary experiment that was only done a live mouse (vs a simulation) for the attention grabbing effect.Obviously an AI agent can perform a simple surgical procedure. The question is how to get it to perform reliably.The reason AI agents work as well as they do with code is because they get feedback and retry in a loop. Surgery is…
-
Very different but made me think of a recent personal story.My father-in-law had an eye accident when he was young, and the doctor at the time said "unfortunately I don't recommend you try surgery, there's a really high risk you go blind".Fast forward to today, and 40 years later such doctor called him and said "hey, turns out we could do a cornea…
-
I don't know what harnesses they run but when I daily tell Claude or others to do X and they indeed do parts of X and then also Y and break A on the way... I'm so tired of asking "but, really?" that I'm not even bothering anymore. The AI folks live in a different world than I live, and whether it's a real world or only in their minds (or only in…
-
From my reading of it, the override moment was because they made a coding error which cause manual adjustment to de-sync the arm position, the agent correctly refused to continue when it noticed that.It does seem like there is minimal "surgery" involved but it's not much of a stretch. If a surgeon took the same steps but someone else opened the…
-
Last night an AI agent performed brain surgery on a mouseLast night an AI agent performed brain surgery on a mouse, yeah'Cause I was sittin' there bored to deathAnd in just 18 hours in front of my laptopIt picked the drugIt picked the brain regionIt drove the stereotactic arms into the brain and injected 250 nanoliters while watching the entry…
-
Every time when I see animal tests or my own meat consumption I wonder how smart all kinds of animals really are. Would I feel worse knowing, I am eating an intelligent being? See its suffering as something more?While I have no moral conundrum about eating meat or doing scientific studies on animals, this seems like a bit unethical.Was it really…
-
For those of us that loathe these kinds of stories being posted on X, this is also being reported on by several outlets.Here's a Wired link:
-
Boy, that's one sketchy looking ESTOP button. I hope the mouse doesn't know it comes in a 2-pack on Amazon for $9.
-
This seems supremely stupid, lazy, and performative. We know how to build custom neural networks to perform much faster with higher accuracy when custom built for a specific task. What was it doing mid-surgery? Googling "brain size of mice"?
-
This is fascinating and scary at the same time. I dread the future when people will have access to cheaper health insurance with the caveat that all consultations and procedures will be performed solely by AI. Want a human? Pay extra.
-
I felt scared for the little guy. Somehow Opus 5 refusing after understanding its limitations felt like the most humane thing in the article.
-
Feels like on boundary of surgery (injecting as opposed to cutting, moving, sowing, etc.) - but maybe I am missing something.
-
The mouse survived!I noticed that the first tweet made no mention if the mouse survived. Seemed a little...sus.
All 2 developments of AI Agent Successfully Injects Gene Therapy Into Live Mouse… →
XHacker NewsGoogle NewsNewswiresMastodon