Welcome to Eye on AI! In this edition\u2026Meta wins AI copyright case in another blow to authors\u2026Google DeepMind releases new AlphaGenome model to better understand the genome…Sam Altman calls Iyo lawsuit \u2018silly\u2019 after OpenAI scrubs Jony Ive deal from website, then shares emails.<\/p>\n
This week, I spoke with Steven Adler, a former OpenAI safety researcher who left the company in January after four years, saying on X after his departure that he was \u201cpretty terrified by the pace of AI development.\u201d Since then, he\u2019s been working as an independent researcher and \u201ctrying to improve public understanding of what the AI future might look like and how to make it go better.\u201d<\/p>\n
What really caught my attention was a new blog post<\/a> from Adler, where he shares his recent experience participating in a five-hour discussion-based simulation, or \u201ctabletop exercise,\u201d with 11 others, which he said was similar to wargames-style exercises in the military and cybersecurity. Together, the group explored how world events might unfold if \u201csuperintelligence,\u201d or AI systems that surpass human intelligence, emerges in the next few years.<\/p>\n A simulation organized by the authors of AI 2027<\/strong><\/p>\n The simulation was organized by the AI Futures Project, a nonprofit AI forecasting group led by Daniel Kokotajlo, Adler\u2019s former OpenAI teammate and friend. The organization drew attention in April for \u201cAI 2027,\u201d a forecast-based scenario mapping out how superhuman AI could emerge by 2027\u2014and what that might mean. According to the scenario, by then AI systems could be using 1,000 times more compute than GPT\u20114 and rapidly accelerating their own development by training other AIs. But this self-improvement could easily outpace our ability to keep them aligned with human values, raising the risk that seemingly helpful AIs might ultimately pursue their own goals.<\/p>\n The purpose of the simulation, said Adler, is to help people understand the dynamics of rapid AI development and what challenges are likely to arise in trying to steer it for the better.<\/p>\n Each participant has their own character whom they try to represent realistically in conversations, negotiations and strategizing, he explained. Those characters included members of the US federal government (each branch, as well as the President and their Chief of Staff), the Chinese government\/AI companies, the Taiwanese government, NATO, the leading Western AI company, the trailing Western AI companies, the corporate AI safety teams, the broader AI safety ecosystem (e.g., METR, Apollo Research), the public\/press, and the AI systems themselves.<\/p>\n Adler was tapped to play what he called \u201cmaybe the most interesting role\u201d\u2014a rogue artificial intelligence. During each 30-minute round of the five-hour simulation, which represented the passage of a few months in the forecast, Adler\u2019s AI got progressively more capable\u2014including at training even more powerful AI systems.<\/p>\n After rolling the dice\u2014an actual, analog pair that was used occasionally in the simulation in cases where it was unclear what would happen\u2014Adler learned that his AI character would not be evil. However, if he had to choose between self-preservation or doing what\u2019s right for humanity, he was meant to choose his own preservation.\u00a0<\/p>\n Then, Adler detailed, with some humor, the awkward interactions his AI character had with the other characters (who asked him for advice on superintelligence), as well as the surprise addition of a second player who played a rogue AI in the hands of the Chinese government.<\/p>\n A power struggle between AI systems<\/strong><\/p>\n The surprise of the simulation, he said, was seeing how the biggest power struggle might not be between humans and AI. Instead, various AIs connecting with each other, vying for victory, might be an even bigger problem. \u201cHow directly AI systems are able to communicate in the future is a really important question,\u201d Adler said. \u201cIt\u2019s really, really important that humans be monitoring notification channels and paying attention to what messages are being passed between the AI agents.\u201d After all, he explained, if AI agents are connected to the internet and permitted to work with each other, there is reason to think they could begin colluding.<\/p>\n Adler pointed out that even soulless computer programs can happen to work in certain ways and have certain tendencies. AI systems, he said, might have different goals that they automatically pursue, and humans need influence over those goals.<\/p>\n The solution, he said, could be a form of AI control based on how cybersecurity professionals deal with \u201cinsider threats\u201d\u2014when someone inside an organization, who has access and knowledge, might try to harm the system or steal information. The goal of security is not to make sure insiders always behave; it\u2019s to build structures that prevent even ill-intentioned insiders from doing serious harm. Instead of just hoping AI systems stay aligned, we should focus on building practical control mechanisms that can contain, supervise, restrict, or shut down powerful AIs\u2014even if they try to resist.<\/p>\n Forecasts and predictions are \u2018hard\u2019<\/strong><\/p>\n I pointed out to Adler that when AI 2027 was released, there was plenty of criticism. People were skeptical, saying the timeline was too aggressive and underestimated real-world limits like hardware, energy, and regulatory bottlenecks. Critics also doubted that AI systems could quickly improve themselves in the runaway way the report suggested and argued that solving AI alignment would likely be much harder and slower. Some also saw the forecast as overly alarmist, warning it could hype fears without solid evidence that superhuman AI is that close.<\/p>\n Adler responded by encouraging others to express interest in running the simulation for their organization (there is a form<\/a> to fill out), but admitted that forecasts and predictions are hard. \u201cI understand why people would feel skeptical, it\u2019s always hard to know what will actually happen in the future,\u201d he said. \u201cAt the same time, from my point of view, this is the clear state of the art in people who\u2019ve sat down and for months done tons of underlying research and interviews with experts and just all sorts of testing and modeling to try to figure out what worlds are realistic.\u201d\u00a0\u00a0<\/p>\n Those experts are not saying that the world depicted in AI 2027 will definitely happen, he emphasized, but \u201cit\u2019s important that the world be ready if it does.\u201d Simulations like this help people to understand what sorts of actions matter and make a difference \u201cif we do find ourselves in that sort of world.\u201d<\/p>\n Conversations with AI researchers like Adler tend to end without much optimism\u2014though it\u2019s worth noting that plenty of others in the field would push back on just how urgent or inevitable this view of the future really is. Still, it\u2019s a relief that his blog post concludes with the hope, at least, that humans will \u201crecognize the challenges and rise to the occasion.\u201d<\/p>\n That includes Sam Altman: If OpenAI hasn\u2019t already run one of these simulations and wanted to try it, said Adler, \u201cI am quite confident that the team would make it happen.\u201d\u00a0<\/p>\n With that, here\u2019s the rest of the AI news.<\/p>\n Sharon Goldman<\/strong> Special Digital Issue: AI at Work
<\/p>\n
<\/p>\n
<\/p>\n
<\/p>\n
<\/p>\n
<\/p>\n
\n<\/ul>\n AI IN THE NEWS<\/p>\n Meta wins AI copyright case in another blow to authors. <\/b>In the same week as a federal judge ruled<\/a> that Anthropic\u2019s use of copyrighted books to train its AI models was \u201cfair use,\u201d Meta also won a copyright case in yet another blow to authors seeking to hold AI companies accountable for using their works without permission. According to the Financial Times<\/a>, Meta\u2019s use of a library millions of books, academic articles and comics to train its Llama AI models was judged \u201cfair\u201d by a federal court on Wednesday. The case was brought by about a dozen authors, including Ta-Nehisi Coates and Richard Kadrey. Meta\u2019s use of these titles is protected under copyright law\u2019s fair use provision, San Francisco district judge Vince Chhabria ruled. Meta had argued that the works had been used to develop a transformative technology, which was fair \u201cirrespective\u201d of how it acquired the works.<\/p>\n Google DeepMind releases new AlphaGenome model to better understand the genome.<\/b> Google DeepMind, the AI research lab famous for developing AlphaGo, the first AI to defeat a world champion Go player, and AlphaFold, which uses AI to predict the 3D structures of proteins, released its new AlphaGenome model<\/a>, designed to analyze up to one million DNA base pairs at once and predict how specific genomic variants affect regulatory functions\u2014such as gene expression, RNA splicing, and protein binding\u2014across diverse cell types. The company said the model was trained on extensive public datasets and achieves state-of-the-art performance on most benchmarks and can assess mutation impacts in seconds. AlphaGenome will be available for non-commercial research, and promises to accelerate discovery in genome biology, disease understanding, and therapeutic development.<\/p>\n Sam Altman calls Iyo lawsuit \u2018silly\u2019 after OpenAI scrubs Jony Ive deal from website, then shares emails. <\/b>On Tuesday, OpenAI CEO Sam Altman on criticized a lawsuit filed by hardware startup Iyo, which accused his company of trademark infringement. CNBC reported <\/a>that in response to the suit,\u00a0 Iyo CEO Jason Rugolo had been \u201cquite persistent in his efforts\u201d to get OpenAI to buy or invest in his company. In a post on X, he wrote that Rugolo is now suing OpenAI over the name in a case he described as \u201csilly, disappointing and wrong.\u201d He then\u00a0posted screenshots of emails on X showing messages between him and Rugolo, which show a mostly friendly exchange.The suit stemmed from an announcement last month that OpenAI was bringing on Apple designer Jony Ive by acquiring his AI startup io in a deal valued at about $6.4 billion. Iyo alleged that OpenAI, Altman and Ive had engaged in unfair competition and trademark infringement and claimed that it\u2019s on the verge of losing its identity because of the deal.\u00a0<\/p>\n FORTUNE ON AI<\/p>\n Can AI help America make stuff again?<\/a> \u2014by Jeremy Kahn\n<\/p>\n AI companies are throwing big money at newly-minted PhDs, sparking fears of an academic \u2018brain drain\u2019<\/a> \u2014by Alexandra Sternlicht<\/p>\n Top e-commerce veteran Julie Bornstein unveils Daydream\u2014an AI-powered shopping agent that\u2019s 25 years in the making<\/a> \u2014by Jason Del Rey<\/p>\n Exclusive: Uber and Palantir alums raise $35M to disrupt corporate recruitment with AI<\/a> \u2014by Beatrice Nolan<\/p>\n AI CALENDAR<\/p>\n July 8-11:\u00a0<\/strong>AI\u00a0for Good Global Summit, Geneva\n<\/p>\n July 13-19:\u00a0<\/strong>International Conference\u00a0on\u00a0Machine Learning (ICML), Vancouver\n<\/p>\n July 22-23:\u00a0<\/strong>Fortune Brainstorm\u00a0AI\u00a0Singapore. Apply to attend\u00a0here<\/a>.\n<\/p>\n July 26-28:\u00a0<\/strong>World\u00a0Artificial\u00a0Intelligence\u00a0Conference (WAIC), Shanghai.\u00a0<\/p>\n Sept. 8-10<\/strong>: Fortune Brainstorm Tech, Park City, Utah. Apply to attend\u00a0here<\/a>.\n<\/p>\n Oct. 6-10:<\/strong>\u00a0World\u00a0AI\u00a0Week, Amsterdam\n<\/p>\n Dec. 2-7:<\/strong>\u00a0NeurIPS, San Diego\n<\/p>\n Dec. 8-9:\u00a0<\/strong>Fortune Brainstorm\u00a0AI\u00a0San Francisco. Apply to attend\u00a0here<\/a>.\n<\/p>\n EYE ON AI NUMBERS130<\/p>\n Many vendors are engaging in \u201cagent washing\u201d\u2014the rebranding of products such as digital assistants, chatbots, and \u201crobotic process automation\u201d (RPA) that either aren\u2019t actually agentic or don\u2019t actually use AI, Gartner says<\/a>, estimating that only about 130 of the thousands of \u201cagentic AI\u201d vendors actually offer real AI agents.<\/p>\n","protected":false},"excerpt":{"rendered":"Welcome to Eye on AI! In this edition\u2026Meta wins AI copyright case in another blow to authors\u2026Google DeepMind…\n","protected":false},"author":3,"featured_media":16857,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[21],"tags":[691,738,16526,305,158,67,132,68],"class_list":{"0":"post-16856","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-eye-on-ai","11":"tag-openai","12":"tag-technology","13":"tag-united-states","14":"tag-unitedstates","15":"tag-us"},"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@us\/114750868906740949","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/16856","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/comments?post=16856"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/16856\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media\/16857"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media?parent=16856"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/categories?post=16856"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/tags?post=16856"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
sharon.goldman@fortune.com<\/a>
@sharongoldman<\/a><\/p>\n
\nFortune recently unveiled a new ongoing series, Fortune AIQ<\/strong><\/a>, dedicated to navigating AI\u2019s real-world impact.\u00a0Our second collection of stories make up a special digital issue of Fortune in which we explore how\u00a0technology is already changing the way the biggest companies do business in finance, law, agriculture, manufacturing, and more.<\/p>\n
<\/p>\n