Anthropic CEO Dario Amodei believes today\u2019s AI models hallucinate, or make things up and present them as if they\u2019re true, at a lower rate than humans do, he said during a press briefing at Anthropic\u2019s first developer event, Code with Claude, in San Francisco on Thursday. <\/p>\n
Amodei said all this in the midst of a larger point he was making: that AI hallucinations are not a limitation on Anthropic\u2019s path to AGI \u2014 AI systems with human-level intelligence or better.<\/p>\n
\u201cIt really depends how you measure it, but I suspect that AI models probably hallucinate less than humans, but they hallucinate in more surprising ways,\u201d Amodei said, responding to TechCrunch\u2019s question.<\/p>\n
Anthropic\u2019s CEO is one of the most bullish leaders in the industry on the prospect of AI models achieving AGI. In a widely circulated paper he wrote last year<\/a>, Amodei said he believed AGI could arrive as soon as 2026. During Thursday\u2019s press briefing, the Anthropic CEO said he was seeing steady progress to that end, noting that \u201cthe water is rising everywhere.\u201d<\/p>\n \u201cEveryone\u2019s always looking for these hard blocks on what [AI] can do,\u201d said Amodei. \u201cThey\u2019re nowhere to be seen. There\u2019s no such thing.\u201d<\/p>\n Other AI leaders believe hallucination presents a large obstacle to achieving AGI. Earlier this week, Google DeepMind CEO Demis Hassabis said today\u2019s AI models have too many \u201choles,\u201d<\/a> and get too many obvious questions wrong. For example, earlier this month, a lawyer representing Anthropic was forced to apologize in court after they used Claude<\/a> to create citations in a court filing, and the AI chatbot hallucinated and got names and titles wrong.<\/p>\n It\u2019s difficult to verify Amodei\u2019s claim, largely because most hallucination benchmarks pit AI models against each other; they don\u2019t compare models to humans. Certain techniques seem to be helping lower hallucination rates, such as giving AI models access to web search. Separately, some AI models, such as OpenAI\u2019s GPT-4.5<\/a>, have notably lower hallucination rates on benchmarks compared to early generations of systems.<\/p>\n However, there\u2019s also evidence to suggest hallucinations are actually getting worse in advanced reasoning AI models. OpenAI\u2019s o3 and o4-mini models have higher hallucination rates than OpenAI\u2019s previous-gen reasoning models<\/a>, and the company doesn\u2019t really understand why.<\/p>\n Later in the press briefing, Amodei pointed out that TV broadcasters, politicians, and humans in all types of professions make mistakes all the time. The fact that AI makes mistakes too is not a knock on its intelligence, according to Amodei. However, Anthropic\u2019s CEO acknowledged the confidence with which AI models present untrue things as facts might be a problem.<\/p>\n In fact, Anthropic has done a fair amount of research on the tendency for AI models to deceive humans, a problem that seemed especially prevalent in the company\u2019s recently launched Claude Opus 4. Apollo Research, a safety institute given early access to test the AI model, found that an early version of Claude Opus 4 exhibited a high tendency to scheme against humans and deceive them<\/a>. Apollo went as far as to suggest Anthropic shouldn\u2019t have released that early model. Anthropic said it came up with some mitigations that appeared to address the issues Apollo raised.<\/p>\n Amodei\u2019s comments suggest that Anthropic may consider an AI model to be AGI, or equal to human-level intelligence, even if it still hallucinates. An AI that hallucinates may fall short of AGI by many people\u2019s definition, though.<\/p>\n","protected":false},"excerpt":{"rendered":"Anthropic CEO Dario Amodei believes today\u2019s AI models hallucinate, or make things up and present them as if…\n","protected":false},"author":2,"featured_media":124114,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3163],"tags":[323,28396,1942,55155,55156,12385,53,16,15],"class_list":{"0":"post-124113","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-anthropic","10":"tag-artificial-intelligence","11":"tag-claude","12":"tag-dario-amodei","13":"tag-hallucinations","14":"tag-technology","15":"tag-uk","16":"tag-united-kingdom"},"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@uk\/114554819217455473","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/posts\/124113","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/comments?post=124113"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/posts\/124113\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/media\/124114"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/media?parent=124113"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/categories?post=124113"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/uk\/wp-json\/wp\/v2\/tags?post=124113"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}