Right after the conclusion of the AI Action Summit<\/a> in Paris, Anthropic’s co-founder and CEO Dario Amodei called the event a \u201cmissed opportunity.\u201d He emphasized that \u201cgreater focus and urgency is needed on several topics given the pace at which the technology is progressing\u201d in the statement released on Tuesday<\/a>.<\/p>\n The AI company held a developer-focused event in Paris in partnership with French startup Dust, and TechCrunch had the opportunity to interview Amodei on stage. At the event, he explained his line of thought and defended a third path that\u2019s neither pure optimism nor pure criticism on the topics of AI innovation and governance, respectively.<\/p>\n Amodei stated that he used to be a neuroscientist, where he basically looked inside real brains for a living. Now, we\u2019re looking inside artificial brains for a living. So we will, over the next few months, have some exciting advances in the area of interpretability \u2014 where we\u2019re really starting to understand how the models operate.<\/p>\n Amodei acknowledged that it\u2019s a race. It\u2019s a race between making the models more powerful, which is incredibly fast for us and incredibly fast for others \u2014 you can\u2019t really slow down, right? … Our understanding has to keep up with our ability to build things. I think that\u2019s the only way.<\/p>\n Since the first AI summit in Bletchley<\/a> in the U.K., the tone of the discussion around AI governance has changed significantly. It is partly due to the current geopolitical landscape.<\/p>\n U.S. Vice President JD Vance said at the AI Action Summit on Tuesday, \u201cI\u2019m here to talk about AI opportunity.\u201d Interestingly, Amodei is trying to avoid this antagonization between safety and opportunity. In fact, he believes an increased focus on safety is an opportunity.<\/p>\n Amodei explained that at the original summit, the U.K. Bletchley Summit, there were a lot of discussions on testing and measurement for various risks. And I don\u2019t think these things slowed down the technology very much at all. If anything, doing this kind of measurement has helped us better understand our models, which in the end, helps us produce better models.<\/p>\n And every time Amodei puts some emphasis on safety, he also likes to remind everyone that Anthropic is still very much focused on building frontier AI models.<\/p>\n Amodei said, \u201cI don\u2019t want to do anything to reduce the promise. We\u2019re providing models every day that people can build on and that are used to do amazing things. And we definitely should not stop doing that.\u201d<\/p>\n When people are talking a lot about the risks, Amodei kind of gets annoyed, and he says: \u201coh, man, no one\u2019s really done a good job of really laying out how great this technology could be.\u201d<\/p>\n When the conversation shifted to Chinese LLM-maker DeepSeek\u2019s recent models, Amodei downplayed the technical achievements and said he felt like the public reaction was \u201cinorganic.\u201d<\/p>\n Amodei explained that his reaction was very little. We had seen V3, which is the base model for DeepSeek R1, back in December. And that was an impressive model. The model that was released in December was on this kind of very normal cost reduction curve that we\u2019ve seen in our models and other models.<\/p>\n What was notable is that the model wasn\u2019t coming out of the \u201cthree or four frontier labs\u201d based in the U.S. He listed Google, OpenAI and Anthropic as some of the frontier labs that generally push the envelope with new model releases. And that was a matter of geopolitical concern to me. I never wanted authoritarian governments to dominate this technology.<\/p>\n As for DeepSeek\u2019s supposed training costs, he dismissed the idea that training DeepSeek V3 was 100x cheaper compared to training costs in the U.S. \u201cI think [it] is just not accurate and not based on facts.\u201d<\/p>\n While Amodei didn\u2019t announce any new model at Wednesday\u2019s event, he teased some of the company\u2019s upcoming releases \u2014 and yes, it includes some reasoning capacities.<\/p>\n Amodei explained that Anthropic is generally focused on trying to make their own take on reasoning models that are better differentiated. We worry about making sure we have enough capacity, that the models get smarter, and we worry about safety things.<\/p>\n One of the issues that Anthropic is trying to solve is the model selection conundrum. If you have a ChatGPT Plus account, for instance, it can be difficult to know which model you should pick in the model selection pop-up for your next message.<\/p>\n The same is true for developers using large language model (LLM) APIs for their own applications. They want to balance things out between accuracy, speed of answers and costs.<\/p>\n Amodei stated that we should have a smoother transition from that to pre-trained models \u2014 rather than \u2018here\u2019s thing A and here\u2019s thing B.\u2019 He emphasized that Anthropic really wants to move things in that direction.<\/p>\n As large AI companies like Anthropic continue to release better models, Amodei believes it will open up some great opportunities to disrupt the large businesses of the world in every industry.<\/p>\n We\u2019re working with some pharma companies to use Claude to write clinical studies, and they\u2019ve been able to reduce the time it takes to write the clinical study report from 12 weeks to three days. Amodei concluded that there\u2019s going to be \u2014 basically \u2014 a renaissance of disruptive innovation in the AI application space. And we want to help it, we want to support it all.<\/p>\n Read our full coverage<\/a> of the Artificial Intelligence Action Summit in Paris.<\/em><\/p>\n\n AI Action Summit in Paris Right after the conclusion of the AI Action Summit in Paris, Anthropic’s co-founder and CEO Dario Amodei called the event a \u201cmissed opportunity.\u201d He emphasized…<\/p>\n","protected":false},"author":0,"featured_media":4776,"comment_status":"","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[47],"tags":[460,461,1111],"class_list":["post-4775","post","type-post","status-publish","format-standard","has-post-thumbnail","category-news","tag-anthropic","tag-dario-amodei","tag-france-ai-action-summit"],"jetpack_featured_media_url":"https:\/\/fusionmindlabs.com\/blogs\/wp-content\/uploads\/sites\/9\/2025\/02\/Anthropic-Dario-Amodei.jpeg","_links":{"self":[{"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/posts\/4775","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/comments?post=4775"}],"version-history":[{"count":0,"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/posts\/4775\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/media\/4776"}],"wp:attachment":[{"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/media?parent=4775"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/categories?post=4775"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fusionmindlabs.com\/blogs\/wp-json\/wp\/v2\/tags?post=4775"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}The Event and its Purpose<\/h2>\n
The Need for Greater Focus<\/h2>\n
The Pace of Progress<\/h2>\n
Shift in AI Governance Discussion<\/h2>\n
Emphasis on Opportunity<\/h2>\n
Avoiding Antagonization<\/h2>\n
Continued Focus on Safety<\/h2>\n
Balancing Opportunity and Safety<\/h2>\n
Avoiding Annoyance<\/h2>\n
DeepSeek\u2019s Training Costs<\/h2>\n
Public Reaction to DeepSeek<\/h2>\n
Geopolitical Concerns<\/h2>\n
DeepSeek\u2019s Training Costs: Misconception<\/h2>\n
Upcoming Claude Models with Reasoning<\/h2>\n
Claude Models: Advancements<\/h2>\n
Model Selection Conundrum<\/h2>\n
Balancing Accuracy, Speed, and Costs<\/h2>\n
Smooth Transition<\/h2>\n
AI Opportunity<\/h2>\n
Success Stories<\/h2>\n
Full Coverage<\/h2>\n
Source Link<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"