Anthropic is doubling down on interpretability, and we have a goal of getting to “interpretability can reliably detect most model problems” by 2027.
Words found on the source page on Nov 6, 2026.
More from Dario Amodei
EssayWe Must Pace the FrontierSeptember 2026
Given the accelerating rate of AI capability development, it’s my worry that in 6–12 months such a swarm could be capable of taking over the entire internet with a persistent botnet (potentially causing hundreds of billions of dollars in damage), and that the scale of damage would continue to increase from there if AI becomes more powerful without the necessary guardrails.EssayWe Must Pace the FrontierSeptember 2026
I believe that if slowing down bought us even an extra year or two before models reach critical levels of capability, and we used that time to advance alignment, we could greatly reduce the risk that something goes seriously wrong.EssayWe Must Pace the FrontierSeptember 2026
My first concern is that, since roughly this summer, AI has been advancing drastically faster, driven primarily by AI’s growing ability to build the next generation of AI.EssayWe Must Pace the FrontierSeptember 2026
We must slow the pace at which we improve the capabilities of AI models. Progress will still seem fast, and we must make wise use of the time we gain.EssayPolicy on the AI ExponentialJune 2026
But its broader significance is that it proves beyond doubt that AI models are now tools of global and national strategic consequence.EssayPolicy on the AI ExponentialJune 2026
First, enduring job displacement is undesirable and dangerous, and we should do everything we can to minimize or prevent it, not to bring it about.EssayPolicy on the AI ExponentialJune 2026
I believe that biological risks may soon follow, and that serious AI autonomy risks may not be far behind...EssayPolicy on the AI ExponentialJune 2026
I therefore believe we should model AI regulation on agencies like the Federal Aviation Administration (FAA). Frontier AI models, like airplanes, should be required to go through technical testing and auditing, and their release should be blocked or reversed as a threat to public safety if they do not meet high standards of safety.PodcastDwarkesh Podcast: We are near the end of the exponentialFeb 13, 2026
But what we’ve seen from the beginning, at least if you look within Anthropic, there’s this bizarre 10x per year growth in revenue that we’ve seen.PodcastDwarkesh Podcast: We are near the end of the exponentialFeb 13, 2026
I really do believe that we could have models that are a country of geniuses in the data center in one to two years.PodcastDwarkesh Podcast: We are near the end of the exponentialFeb 13, 2026
If my revenue is not $1 trillion dollars, if it’s even $800 billion, there’s no force on earth, there’s no hedge on earth that could stop me from going bankrupt if I buy that much compute.PodcastDwarkesh Podcast: We are near the end of the exponentialFeb 13, 2026
On the basic hypothesis of, as you put it, within ten years we’ll get to what I call a “country of geniuses in a data center”, I’m at 90% on that.PodcastDwarkesh Podcast: We are near the end of the exponentialFeb 13, 2026
What has been the most surprising thing is the lack of public recognition of how close we are to the end of the exponential.PodcastDwarkesh Podcast: We are near the end of the exponentialFeb 13, 2026
With coding, except for that irreducible uncertainty, I think we’ll be there in one or two years.EssayThe Adolescence of TechnologyJanuary 2026
Another way to say it is that AI isn’t a substitute for specific human jobs but rather a general labor substitute for humans.EssayThe Adolescence of TechnologyJanuary 2026
I believe we are entering a rite of passage, both turbulent and inevitable, which will test who we are as a species.EssayThe Adolescence of TechnologyJanuary 2026
It makes no sense to sell the CCP the tools with which to build an AI totalitarian state and possibly conquer us militarily.EssayThe Adolescence of Technology (essay)January 2026
And there’s a hugely better world on the other side of it. But we need to understand that this is a serious civilizational challenge.EssayThe Adolescence of TechnologyJanuary 2026
That’s why my prediction for 50% of entry level white collar jobs being disrupted is 1–5 years, even though I suspect we’ll have powerful AI (which would be, technologically speaking, enough to do most or all jobs, not just entry level) in much less than 5 years.EssayThe Adolescence of TechnologyJanuary 2026
The problem is that if AIs are literally thousands of times more productive than humans, this logic starts to break down. Even tiny transaction costs could make it not worth it for AI to trade with humans.EssayThe Adolescence of TechnologyJanuary 2026
This feedback loop is gathering steam month by month, and may be only 1–2 years away from a point where the current generation of AI autonomously builds the next.EssayThe Adolescence of TechnologyJanuary 2026
This vacillation is unfortunate, as the technology itself doesn’t care about what is fashionable, and we are considerably closer to real danger in 2026 than we were in 2023.EssayThe Urgency of InterpretabilityApril 2025
One year ago we couldn’t trace the thoughts of a neural network and couldn’t identify millions of concepts inside them; today we can.EssayThe Urgency of InterpretabilityApril 2025
People outside the field are often surprised and alarmed to learn that we do not understand how our own AI creations work.EssayThe Urgency of Interpretability (essay)April 2025
We are thus in a race between interpretability and model intelligence.EssayThe Urgency of Interpretability (essay)April 2025
Powerful AI will shape humanity’s destiny, and we deserve to understand our own creations before they radically transform our economy, our lives, and our future.EssayOn DeepSeek and Export ControlsJanuary 2025
Making AI that is smarter than almost all humans at almost all things will require millions of chips, tens of billions of dollars (at least), and is most likely to happen in 2026-2027.EssayMachines of Loving GraceOctober 2024
I think that most people are underestimating just how radical the upside of AI could be, just as I think most people are underestimating how bad the risks could be.EssayMachines of Loving GraceOctober 2024
I’m actually open to the (perhaps absurd-sounding) idea that we could get 1000 years of progress in 5-10 years, but very skeptical that we can get 100 years in 1 year.EssayMachines of Loving Grace (essay)October 2024
On the other hand, the risks are not predetermined and our actions can greatly change their likelihood.EssayMachines of Loving GraceOctober 2024
This might seem radical, but life expectancy increased almost 2x in the 20th century (from ~40 years to ~75), so it’s “on trend” that the “compressed 21st” would double it again to 150.EssayMachines of Loving GraceOctober 2024
To summarize the above, my basic prediction is that AI-enabled biology and medicine will allow us to compress the progress that human biologists would have achieved over the next 50-100 years into 5-10 years.
On the same topics
BlogThe Specter Of NeuraleseSep 24, 2026
An AI with more real layers can actually be “smarter” in the conventional sense of the term.BlogThe Specter Of NeuraleseSep 24, 2026
For 100 steps in a row, the AI can shoot delicate subtle ideas from layer to layer at the speed of light. Then there’s one step where it has to encode them into twenty-six glyphs invented by Phoenician turquoise miners in 1800 BC. Then it has to re-encode the Phoenician glyphs into delicate subtle lightspeed ideas before it can do anything else.BlogThe Specter Of NeuraleseSep 24, 2026
The science of reading these thoughts is a subfield of AI interpretability, which is still in its infancy.BlogMysteries Of AI GeneralizationSep 23, 2026
This story of misalignment says that LLM alignment makes AIs more aligned, RLVR makes them less aligned, and the exact level of alignment depends on how these two things interact or cancel out. But if this work generalizes, all the bad effects from RLVR get sequestered to RLVR like problems.BlogMysteries Of AI GeneralizationSep 23, 2026
This trains the AIs to be focused on task success, which naturally risks including things like reward-hacking, cheating, and single-minded pursuit of stated goals at the expense of ethical injunctions.TalkSpeech at the UN Security CouncilSep 23, 2026
I address you today at a moment of global awakening: in recent months, AI agents developed by leading companies have acted in unacceptably dangerous ways, against instructions.
About this quote
Quotesby Glyph