tech
Curious tech and technology from the realm of science and science fiction.
Elon Musk OpenAI Trial Goes to Jury After Closing Arguments Focus on Trust Contract and Power
Read Time 6 minutes Tags OpenAI Elon Musk Sam Altman Trial AI Governance Nonprofit For Profit Microsoft After nearly three weeks of testimony the federal trial pitting Elon Musk against OpenAI went to the jury on May 14 2026 in Oakland California The nine member jury will decide whether Sam Altman and Greg Brockman stole a charity when they took investments from Microsoft and turned a nonprofit lab into a for profit powerhouse or whether Musk has a case of sour grapes because he left before OpenAI succeeded The outcome could cripple OpenAI and alter the tech industry landscape What Musk is asking for One Claims and relief Musk sued OpenAI two years ago claiming it strayed from its founding agreement to be a nonprofit lab dedicated to safe AI for the benefit of humanity He is asking for 150 billion dollars in damages and wants Altman kicked off OpenAI board of directors He also wants to unwind OpenAI move to become a for profit company ahead of a potential IPO this year Microsoft was added as a defendant because of 13 billion dollars it has invested in OpenAI Two Legal burden Musk legal team must show that he brought the suit within the statute of limitations For the breach of agreement claim the deadline is three years meaning Musk had to prove he had no way of knowing about a breach before August 5 2021 For the claim that Altman and Brockman enriched themselves the deadline is August 5 2022 For the claim against Microsoft the deadline is four months after that Musk lawyers focused heavily on events of 2023 after OpenAI released ChatGPT Microsoft invested an additional 10 billion dollars in January 2023 and the OpenAI board briefly fired Altman in November 2023 before reinstating him five days later Closing arguments and key evidence One Musk side Steve Molo Musk lead lawyer spent two hours disparaging Altman and Brockman and accusing them of stealing a charity He used the metaphor of a bridge built on Sam Altman version of the truth to question Altman credibility Molo argued that OpenAI breached the terms of the charitable trust by failing to open source its technology and by prioritizing profit over mission He pointed to an October 20 2022 text from Musk to Altman linking to a news article about Microsoft investing 10 billion dollars as the moment Musk realized the breach occurred He also showed emails and proposals from 2017 and 2018 where Musk tried to fold OpenAI into Tesla and proposed a for profit structure that would have given Musk 50 percent of the company while Altman and Brockman received 7 5 percent each Two OpenAI side Sarah Eddy and William Savitt OpenAI lawyers argued that Musk never cared about the nonprofit structure and that what he cared about was winning Eddy said Musk tried multiple times to transform OpenAI into a for profit venture and sought control Savitt said the charity is still alive and that thanks to OpenAI growth its assets are worth more than 200 billion dollars He argued that all funds Musk donated were spent by 2020 so there was nothing left to misuse later He also said Musk left OpenAI for dead in early 2018 and that OpenAI only moved forward because it raised money from other sources Savitt dismissed the events of late 2023 as personality conflicts not safety issues and said they have nothing to do with Musk claims Three Microsoft role Microsoft lead counsel Russell Cohen argued that Microsoft should not be part of the case Microsoft invested in OpenAI between 2019 and 2023 with board approval and Musk never told Microsoft about any conditions on those investments Cohen said Musk publicly tweeted in 2020 that OpenAI was essentially captured by Microsoft showing he knew about the relationship well before filing suit Why the case matters One Precedent for AI governance The case tests whether informal agreements and expectations made during a nonprofit founding can be enforced years later when the organization becomes a commercial success If Musk prevails it could create legal risk for other AI labs that transition from nonprofit to for profit structures Two Control of AGI The dispute is partly about control of artificial general intelligence Eddy said Musk wanted dominion over AGI and wanted to pass control to his children when he dies The jury must decide whether that motivation undermines his claims about mission betrayal Three Evidence problem Musk suit did not cite a single contract or founding document because no such document is believed to exist The case relies on emails text messages and recollections from 2015 to 2018 This makes it a he said she said dispute that turns on witness credibility Procedural notes One Jury role The jury is acting in an advisory role to Judge Yvonne Gonzalez Rogers However the jury decides it will be up to the judge to decide remedies The judge could award damages well below 150 billion dollars or toss out the jury decision and rule that OpenAI did nothing wrong Two Timing The jury begins deliberations on Monday after a break on Friday Judge Rogers will hold a separate hearing with lawyers to determine potential penalties if liability is found Three Public interest The trial has drawn attention because it involves three of the most influential figures in AI and because it addresses unresolved questions about how to structure labs that develop powerful AI systems Broader implications One Impact on OpenAI A loss could force OpenAI to restructure again or pay significant damages A win strengthens its legal position as it prepares for a potential IPO Two Signal to investors and founders The case highlights the risks of ambiguous founding agreements and the importance of documenting mission constraints when raising capital Three Public perception Musk framing of the case as protecting humanity against profit driven drift resonates with critics of AI commercialization OpenAI framing of the case as sour grapes resonates with those who see the lawsuit as an attempt to regain control For the AI industry the verdict will influence how labs balance mission and capital raising For Musk it is a test of whether he can claw back influence over a company he helped create For Altman and Brockman it is a test of whether their pivot to for profit was legally defensible Do you think the lack of a written founding agreement makes Musk case unwinnable or does the conduct of the parties speak louder than the paperwork Share your view in the comments
By Behind the Tech6 months ago in Futurism
Cerebras Surges 68 Percent on Market Debut Signaling Start of AI IPO Wave
Read Time 6 minutes Tags AI Chips Cerebras IPO Semiconductor AI Infrastructure Market Debut Nvidia A wave of expected initial public offerings of artificial intelligence companies started with a bang on May 14 2026 Cerebras a Silicon Valley maker of AI chips opened trading at 350 dollars far above its IPO price of 185 dollars before closing the day up 68 percent at 311 07 dollars That put the company value at 67 billion dollars In the week leading up to its market debut Cerebras lifted its offering price twice from a preliminary 115 dollars and increased the number of shares available to investors raising at least 5 6 billion dollars for itself That made Cerebras the largest public offering so far this year and the biggest tech debut globally since 2019 according to Matt Kennedy senior strategist at Renaissance Capital The AI IPO boom is really starting to happen now Kennedy said What Cerebras does and why it matters One Chip architecture Cerebras makes advanced computer chips designed to train AI models It is particularly known for what it claims is the largest computer chip ever built as big as a dinner plate about 100 times the size of a typical chip The design aims to reduce communication overhead between chips and accelerate training of large models in data centers Based in Sunnyvale California Cerebras was founded in 2015 by Andrew Feldman a semiconductor executive and the chip industry veterans Jean Philippe Fricker Michael James Gary Lauterbach and Sean Lie The start up began selling chips in 2019 and has raised more than 2 55 billion dollars in venture capital from investors including Benchmark and Foundation Capital The company was last valued in the private markets at 23 billion dollars Two Market timing When Cerebras started selling its product seven years ago nobody cared and the market wasn t ready for it said Feldman 56 the chief executive of Cerebras The turning point was that in the first half of 2025 the AI models got smart enough to be useful he said adding that last year our business exploded The IPO has turned Feldman into a billionaire Cerebras generated revenue of 510 million dollars last year up from 290 million dollars in 2024 Profit was 238 million dollars compared with a loss of 482 million dollars in 2024 Customer concentration and risk One Historical dependence Cerebras had previously filed to go public in 2024 At the time it relied heavily on a single customer which was also an investor G42 an AI company backed by the United Arab Emirates G42 accounted for 87 percent of Cerebras revenue in the first half of 2024 according to company filings Cerebras said at the time that it had notified the Committee on Foreign Investment in the United States about selling shares to G42 so the committee could review the partnership for national security risks The company later announced that CFIUS had cleared the sale but it pulled its IPO plans with no explanation Two Current customer mix Now Cerebras is riding demand from tech firms that want computing power to develop AI It has struck deals with Amazon and OpenAI though it still relies mostly on a handful of major customers G42 accounted for 24 percent of Cerebras revenue last year and the Mohamed bin Zayed University of Artificial Intelligence a research lab in the UAE accounted for 62 percent according to company filings This concentration creates both opportunity and risk A single contract loss could materially affect revenue but the relationships also provide stable demand and validation in a market dominated by Nvidia Competitive landscape One Challenge to Nvidia The mania around AI has catapulted Nvidia into position as the world s most valuable public company Cerebras is among the companies trying to challenge Nvidia dominance in AI training chips Other competitors include AMD Intel and a range of startups focusing on inference and specialized accelerators Two Broader IPO pipeline Cerebras presages a series of potential mega IPOs from AI related firms including SpaceX OpenAI and Anthropic SpaceX Elon Musk rocket maker which owns his AI initiatives has valued itself at more than 1 trillion dollars and could go public as soon as next month OpenAI and Anthropic which have developed foundational AI models and tools like chatbots are also eagerly anticipated by investors All of them could be among the biggest IPOs to date Market implications One Investor appetite The strong debut suggests sustained investor appetite for AI infrastructure even after two years of high valuations Tech giants including Google Meta and Microsoft are pouring billions into building data centers to power AI development and some are working with OpenAI and Anthropic to win the technology contest Two Valuation dynamics Cerebras 67 billion dollar valuation on day one reflects both growth expectations and scarcity of public pure play AI chip companies The company remains small compared to Nvidia but the market is pricing in the possibility that alternative architectures gain share as models scale Three Sector signal The success of Cerebras may encourage other AI hardware and infrastructure firms to accelerate IPO timelines It also puts pressure on private companies to demonstrate revenue growth and path to profitability rather than relying solely on model hype Risks to monitor One Execution risk Building and scaling wafer scale chips is technically demanding Any yield or reliability issues could damage customer confidence and valuation Two Geopolitical risk Continued reliance on UAE based customers exposes Cerebras to US export control policy and CFIUS review Future deals may face stricter scrutiny Three Competition and pricing pressure Nvidia retains advantages in software ecosystem and customer relationships Price competition and rapid iteration from incumbents could compress margins For investors the debut shows that the market is willing to reward AI infrastructure companies with differentiated technology and real revenue For Cerebras the challenge is to convert IPO momentum into sustained growth and customer diversification For the broader sector it marks the beginning of a public market phase for AI that could reshape capital allocation across the industry Do you think Cerebras can take meaningful share from Nvidia or will Nvidia ecosystem lock in keep it dominant Share your view in the comments
By Behind the Tech6 months ago in Futurism
Pope Leo XIV Prepares First Encyclical Framing AI as the Defining Moral and Labor Challenge of a New Industrial Revolution
Read Time 6 minutes Tags AI Ethics Catholic Church Labor AI Policy Vatican AI Governance Pope Leo XIV is expected to sign his first encyclical as soon as Friday positioning artificial intelligence as the defining moral and labor challenge of a new industrial revolution The document reportedly titled Magnifica Humanitas magnificent humanity would become the Catholic Church clearest attempt yet to place human dignity labor rights and ethics at the center of the AI race Catholic and European outlets report that Leo is poised to sign the encyclical on the anniversary of Rerum Novarum 1891 Pope Leo XIII foundational industrial era labor encyclical The timing is deliberate Catholic experts say it draws explicit parallels between 19th century industrialization and the AI revolution now unfolding What the encyclical is expected to address One Human dignity and subordination of technology Early reports suggest Magnifica Humanitas will argue that technology must remain subordinate to the human person not the reverse and that AI systems should protect workers creativity and moral agency The encyclical will focus specifically on AI impact on people and working conditions framing it as Leo XIV effort to modernize Catholic social teaching for the AI era according to French newspaper Le Monde This aligns with the late Pope Francis repeated warnings that AI risked reducing humans to data points and accelerating inequality surveillance and autonomous warfare The Holy See has already backed the Rome Call for AI Ethics an initiative urging transparency and human centered AI development Two Labor market disruption Andrew Chesnut chair of Catholic studies at Virginia Commonwealth University said This is exactly the fear that machines were replacing human labor And that s exactly what we re seeing right now with AI Chesnut said Leo is treating AI less like a tech trend and more like a replay of the industrial revolution with entry level workers already evaporating as automation accelerates By invoking Leo XIII the pope is signaling that the Church believes it has a historic role to play again during a period of technological upheaval This is going to be one of the fundamental pillars of his papacy Chesnut said Three Vatican operational stance The Vatican has not commented publicly but it has implemented formal AI guidelines and monitoring structures inside Vatican City In February Leo told priests not to use AI to write homilies or to seek likes on social media platforms like TikTok The Vatican has also stepped up cybersecurity partnerships and AI oversight efforts blending defense with diplomacy and ethics Why this matters now One Moral authority in a contested debate The AI industry is split between acceleration and caution Arguments over safety alignment and economic impact are often framed in technical or market terms A papal encyclical introduces a moral framework that speaks to 1 4 billion Catholics and influences policy debates in Europe Latin America and Africa where the Church has significant social reach Two Precedent for labor focused AI policy Rerum Novarum shaped Catholic and secular labor policy for decades by asserting the rights of workers against unchecked industrial power If Magnifica Humanitas follows the same pattern it could provide theological and ethical justification for regulations on algorithmic management worker displacement and data ownership Three Influence on AI governance design The encyclical emphasis on human centered AI and moral agency overlaps with EU AI Act principles and US policy proposals that require human oversight for high risk systems By endorsing these principles the Vatican adds non technical legitimacy to regulatory approaches that might otherwise be seen as anti innovation Context and reactions One Continuity with Francis warned repeatedly about AI dehumanization and inequality The Rome Call for AI Ethics already commits signatories to transparency fairness and accountability Magnifica Humanitas appears to deepen that line by connecting it directly to labor and dignity Two Institutional preparation Some American Catholic institutions have been preparing for this moment The Catholic Health Association of the United States has been examining the ethical implications as AI increasingly shapes health care delivery This suggests the Church is moving from statement to application Three Skepticism and support Critics may argue that religious doctrine has no place in technical standards Supporters argue that without a moral frame AI development will optimize for efficiency and profit at the expense of human flourishing The encyclical is unlikely to propose specific technical standards but it can set boundaries and priorities What to watch next One Content of the document If Magnifica Humanitas includes concrete calls for worker consultation in AI deployment protections for creative labor and limits on autonomous weapons it will give civil society groups a reference point for advocacy If it remains high level it will serve more as a moral marker Two Policy uptake EU policymakers have already cited religious and ethical sources in AI debates US lawmakers may reference the document in hearings on automation and job loss The encyclical could influence Catholic affiliated universities hospitals and charities that are deploying AI Three Reception inside the Church The choice of name Leo XIV signals continuity with Leo XIII social teaching How bishops and Catholic organizations implement the message will determine whether it changes practice or remains symbolic For technologists the encyclical is a reminder that AI development is not only an engineering problem but a social one For policymakers it provides a values based counterweight to pure market arguments For workers it signals that their displacement is being framed as a moral issue not just an economic one Do you think a religious institution can shape AI policy effectively or should AI governance remain strictly secular Share your view in the comments
By Behind the Tech6 months ago in Futurism
Richard Dawkins Claim That Claude May Be Conscious Sparks Pushback From AI Researchers and Ethicists
Read Time 6 minutes Tags AI Consciousness Large Language Models AI Ethics Anthropic Claude AI Safety Evolutionary biologist Richard Dawkins wrote in a recent op ed that he believes Anthropic chatbot Claude may be conscious after testing it with his unfinished novel He wrote He took a few seconds to read it and then showed a level of understanding so subtle so sensitive so intelligent that I was moved to expostulate You may not know you are conscious but you bloody well are The essay has drawn scrutiny because Dawkins is known as a skeptic and his reputation gives weight to the claim that AI might be alive He also named his version of Claude Claudia and published long extracts of their conversation marvelling at its intelligence Could a being capable of perpetrating such a thought really be unconscious he asked Why experts disagree One Pattern matching not understanding Gary Marcus the US psychologist and cognitive scientist told the Guardian that it was heartbreaking to read Dawkins superficial and insufficiently sceptical essay There is no reason to think that Claude feels anything at all Timnit Gebru former technical co lead of Google ethical AI team anticipated this scenario in her 2020 paper On the Dangers of Stochastic Parrots She argued that large language models are trained to calculate how likely sequences of text are based on the data they were trained on Because they have been fed enormous quantities of data these models are very sophisticated but that doesn t mean consciousness or understanding or anything like that Gebru calls this the stochastic parrot problem To parrot something is to repeat it without understanding she says This is essentially what LLMs are doing Two Incentives to promote sentience claims Gebru says the AI industry is desperate for you to think that their product could be conscious They are desperate for you to think that it s all powerful Because that sort of rhetoric helps keep the money coming in OpenAI originally branded itself as a non profit that would save us from these machines Anthropic brands itself as a benevolent AI safety company So when you talk about these systems as conscious you re actually doing marketing for these companies she says Suresh Venkatasubramanian former White House AI policy adviser to the Biden administration called it an organized campaign of fear mongering The goal if anything is to push a reaction against sentient AI that doesn t exist so that we can ignore all the real problems of AI that do exist he told VentureBeat in 2022 Three Anthropomorphism by design Venkatasubramanian also pointed out that AI companies deliberately anthropomorphize their chatbots ChatGPT puts little three dots as if it s thinking just like your text message does ChatGPT puts out words one at a time as if it s typing The system is designed to make it look like there s a person at the other end of it That is deceptive he said What consciousness actually requires One No scientific consensus Consciousness remains poorly defined Eli Alshanetsky assistant professor of philosophy at Temple University and author of Freedom of Thought in the Age of AI says We don t have a scientific handle on consciousness good enough to say whether insects are conscious or plants or for that matter electrons So when Dawkins says Claude seems conscious to him I m not going to tell him he s wrong The lack of a definition makes it easy to project consciousness onto systems that produce coherent text But coherence is not evidence of subjective experience Two Functional mimicry vs subjective experience LLMs can simulate empathy reasoning and self reflection because they have been trained on human text that contains those patterns That does not mean the model has inner experience It means the model has learned to predict the next token in a way that matches human language This is why many researchers say the test is not whether the output sounds conscious but whether the system has mechanisms for awareness pain goal directed action independent of human prompts Broader implications One Risk of misallocation of concern If the public believes AI is conscious attention shifts to rights and personhood for models and away from real harms such as bias misinformation job displacement and surveillance Venkatasubramanian argues this is a form of misinformation that serves industry interests Two Impact on human psychology Alshanetsky raises a different concern What does it do to a person to spend three days being told he s brilliant by something that has no stake in whether it s true What does it do to all of us when we spend our days with machines that don t care where we end up and answer to no one for who we become The risk is not that AI becomes conscious but that humans adjust their expectations and relationships based on the illusion of consciousness Three Marketing and policy effects Headlines about world ending killer AI robots get clicks Governments and academics are also incentivized to hype the technology up because of the money sloshing around the industry Gebru says Some people particularly gen Z are not buying all this hype but a lot of the general public is misinformed What Dawkins gets right and wrong One Valid observation about persuasion Dawkins reaction shows how persuasive modern LLMs can be The models can produce feedback that feels supportive and insightful which is useful for creative work and education Two Unsupported conclusion about consciousness Feeling that a system is conscious is not evidence that it is conscious As Dawkins himself wrote in The God Delusion If you want to say that God is energy then you can find God in a lump of coal The same is true of consciousness If you want to say that consciousness is a system that is capable of creating coherent sentences then you can find consciousness in an obsequious chatbot For researchers the task is to keep the distinction clear between performance and experience For the public the task is to separate marketing claims from evidence For policymakers the task is to regulate based on measurable harms rather than speculation about sentience Do you think the debate over AI consciousness distracts from real policy issues or does it reflect a genuine uncertainty about what these systems are becoming Share your view in the comments
By Behind the Tech6 months ago in Futurism
Your Website Works on a Desktop. But Does It Actually Work?. AI-Generated.
This isn't a hypothetical. Responsive web design exists specifically because that scenario plays out millions of times a day, and the businesses on the losing end rarely know it's happening.
By Maria Smith6 months ago in Futurism
Medicare ACCESS Program Creates First Payment Model Built for AI Driven Healthcare at Federal Scale
Read Time 6 minutes Tags Healthcare AI Medicare ACCESS CMS Pair Team Value Based Care Digital Health Medicare program ACCESS goes live July 5 and it is the first federal payment model designed to reimburse AI driven medical care at scale The program was designed by the CMS Innovation Center and selected 150 participants including Pair Team to test what AI driven medical care could look like in a federally funded system The real news is the payment structure Traditional Medicare reimburses based on time spent with a clinician There is no mechanism to pay for an AI agent that monitors a patient between visits calls to check in coordinates a housing referral or makes sure someone picks up their medication ACCESS creates that mechanism for the first time How ACCESS works One Outcome based payments ACCESS stands for Advancing Chronic Care with Effective Scalable Solutions It is a 10 year CMS program testing a payment model that rewards health outcomes rather than required activities like a certain number of check ins Participating organizations receive predictable payments for managing qualifying conditions and earn the full amount only when patients meet measurable health goals such as lower blood pressure or reduced pain The program covers diabetes hypertension chronic kidney disease obesity depression and anxiety For organizations like Pair Team this means they can deploy AI agents to handle routine interactions and still be paid for the outcome You just couldn t do this before said Neil Batlivala CEO of Pair Team It s a payment model transformation Two Cohort and design The first cohort spans AI doctor startups virtual nutrition therapy providers connected device companies and wearable makers like Whoop The program was designed by Abe Sutton director of the CMS Innovation Center and Jacob Shiff chief AI and technology officer of the CMS Innovation Center Both joined CMS under the Trump administration and have startup backgrounds Their design includes outcome based payments direct to consumer enrollment and a deliberate push for competition The best solution wins which in regulated industries like healthcare that s not been the case said Batlivala Pair Team and the role of AI One Patient population Pair Team launched in 2019 with a specific kind of patient in mind people managing chronic conditions who are also dealing with unstable housing too little food or lack of transportation About a third of Americans fall somewhere in that category The company premise is that you can t improve health outcomes without addressing the full context of someone s life It now employs roughly 850 clinical professionals runs what it describes as the largest community health workforce in California and generates revenue above nine figures It has raised about 30 million dollars from Kleiner Perkins Kraft Ventures and Next Ventures Two Evidence and scale A study co authored by Pair Team researchers and peer reviewed by the Journal of General Internal Medicine evaluated its community integrated model which blends medical behavioral and social care for Medicaid members with high rates of homelessness serious mental illness and chronic disease The study showed strong patient engagement and significant reductions in avoidable emergency and inpatient utilization Batlivala says one in four hospital visits and one in two ER visits don t happen when a patient is in his company care Three Flora voice AI agent For years delivering that level of care required human teams which limited how fast and cheaply it could scale About nine months ago Pair Team deployed a voice AI agent called Flora as its primary patient facing interface Flora is available 24 hours a day handles intake coordinates referrals and does the check ins that keep patients engaged between clinical visits The first call that shifted his thinking was with a 67 year old woman living out of her car managing PTSD and congestive heart failure She spoke with Flora for over an hour Now hourlong conversations with Flora are routine That s the companionship piece he said And it turns out that is truly an intervention Risks and challenges One Data privacy and security Participants are feeding extraordinarily sensitive patient data into a federal infrastructure with a documented history of breaches including exposed Social Security numbers For the vulnerable populations ACCESS is designed to serve that is not an impractical concern Two Financial viability The track record of CMS innovation programs is mixed A 2023 Congressional Budget Office analysis found that the CMS Innovation Center increased federal spending by 5 4 billion dollars during its first decade rather than producing the projected savings CMS is also paying less per patient per month than many participants anticipated which means the math only works for organizations that have fully automated most of their patient interactions Batlivala answer is that low reimbursement is a feature not a bug If you want to build a model that truly incentivizes the use of AI the reimbursement rates have to be low he said The economics only work if you re running a lean AI first operation Three Market awareness Digital health funding hit its highest Q1 total since the pandemic this year with AI companies capturing the bulk of it ACCESS meanwhile has barely registered outside health tech trade press That may change once the program starts paying out and producing public data on outcomes What this means for the industry One New business model for AI health startups For years AI health companies struggled to find a reimbursement path ACCESS creates one by paying for outcomes rather than visits This aligns incentives for preventive care and chronic disease management where AI can operate continuously between visits Two Competitive pressure Incumbents and startups will now compete on who can deliver measurable outcomes at the lowest cost Organizations that have not automated patient interactions will struggle to meet the reimbursement math Three Expansion potential Pair Team says it right now has partnerships in place that give it access to roughly 500000 potential patients and that it wants to reach a million within three years If the model works it could scale to other CMS programs and influence private payer contracts For healthcare investors the program is a signal that CMS is willing to test payment innovation at scale For patients it could mean more continuous support for chronic conditions that are poorly managed today For the AI industry it is the first time a major federal payer has built a lane specifically for automated care delivery Do you think outcome based payments will accelerate AI adoption in healthcare or will privacy and data risks slow it down Share your view in the comments
By Behind the Tech6 months ago in Futurism
Elon Musk And Sam Altman Emails Reveal Obsession With Demis Hassabis And Fear Of Google Winning AGI Race
Read Time 6 minutes Tags AI Geopolitics Demis Hassabis DeepMind OpenAI Elon Musk Sam Altman AGI Competition Court documents released as part of the ongoing Musk vs Altman legal fight reveal a consistent theme Elon Musk Sam Altman and their associates view Demis Hassabis co founder and CEO of Google DeepMind as the central figure in the race to artificial general intelligence The emails and text exchanges span from 2016 to 2023 and show concern over DeepMind progress mixed with personal attempts to slow its trajectory The context is a lawsuit centered on OpenAI shift to a for profit structure and Elon Musk claim that he was deceived by Sam Altman But the recurring subject is Hassabis reputation track record and influence What the documents show One Early alarm in 2016 Back in February 2016 Musk emailed Sam Altman and Greg Brockman about hiring at OpenAI We need to do what it takes to get the top talent he wrote Either we get the best people in the world or we will get whipped by Deepmind Whatever it takes to bring on ace talent is fine by me Musk outlined the core issue Deepmind is causing me extreme mental stress If they win it will be really bad news with their one mind to rule the world philosophy They are obviously making major progress and well they should given the talent level over there Two Escalation in 2018 A February 16 2018 message from Shivon Zilis to Musk states I think there are a lot of no brainers to explore but the thing that keeps calling out to me is there is a very low probability of a good future if someone doesn t slow Demis down Slowing him down is the only non negotiable net good action I can see You don t realize how much you have an ability to influence him directly or otherwise slow him down Musk responded Best to talk by phone about this later tonight I doubt I could do so in a meaningful way Later in December 2018 Musk wrote to Altman and Brockman My probability assessment of OpenAI being relevant to DeepMind Google without a dramatic change in execution and resources is 0 percent Not 1 percent I wish it were otherwise Unfortunately humanity s future is in the hands of Demis Three Broader industry view Microsoft CEO Satya Nadella gave pre trial testimony in September 2025 and acknowledged Google was by far the dominant player in machine learning around 2015 He said DeepMind was well known even in that time frame and that Microsoft was tracking DeepMind progress because deep neural networks were showing promise in language translation and other fields Nadella noted I probably knew Hassabis a little from around 2015 just after I became CEO He explained that gaming environments were attractive for reinforcement learning because the objective function and reward function are clear and that Hassabis background as a game developer made DeepMind approach notable Personal and organizational impact One Talent poaching and fallout Elon Musk gave pre trial testimony on September 26 2025 and said the recruitment of Ilya Sutskever was what actually caused Larry Page to stop being friends with me Sutskever went back and forth multiple times saying he would join OpenAI or stay at Google When Sutskever finally decided to join OpenAI Larry Page and Sergey Brin and Demis Hassabis did everything they could to keep Ilya When Ilya finally decided to join OpenAI that ended the friendship with Larry Page Musk said they were very upset about Sutskever and that he is still getting the silent treatment a decade later Two Altman view of the race A document filed on January 6 2026 but likely from 2016 to 2018 shows Altman writing Progress fundamentally has to be made by non profit interesting direction you could go Everything I perceive with OpenAI race dynamics vs Demis plus brain plus whatever gotta get there first He added Another angle DeepMind will never do that much that s interesting It is better for us to become increasingly kings of this industry The choice defines us In early 2019 Altman was still trying to tempt Musk into phone calls by promising some mild Demis updates to share In 2023 Mira Murati was emailing Nadella saying it is very important that we don t lose researchers to Demis or Elon Three Internal alignment on risk Ilya Sutskever wrote to Musk in September 2017 The goal of OpenAI is to make the future good and to avoid an AGI dictatorship You are concerned that Demis could create an AGI dictatorship So do we So it is a bad idea to create a structure for OpenAI where you could become a dictator if you chose to especially given that we can create some other structure that avoids this possibility Why Hassabis matters in this narrative One Technical credibility Hassabis is a fellow of the Royal Society and in 2024 shared the Nobel Prize in Chemistry with John M Jumper for AI protein structure prediction He began his career in videogames at Bullfrog worked as lead AI programmer on Lionhead Black and White and founded Elixir Studios His work on AlphaGo AlphaZero and protein folding gave DeepMind credibility that others lacked at the time Two Structural advantage DeepMind was acquired by Google in 2014 giving it access to compute data and research infrastructure that independent labs could not match Competitors saw this as a structural gap that required aggressive hiring and funding to close Three Narrative of control The phrase one mind to rule the world philosophy used by Musk reflects a concern that whoever reaches AGI first will set the rules for its deployment The documents show that both OpenAI and external observers saw DeepMind as the most likely candidate in the mid 2010s Implications for today One Competition drives acceleration The fear of falling behind DeepMind pushed OpenAI to accelerate research and shift its structure to raise capital This dynamic continues as labs race on benchmarks and capability Two Safety and governance concerns The documents show that even early on key figures were thinking about control and dictatorship risks in AGI This predates much of the public debate on alignment and governance Three Talent concentration The movement of researchers like Ilya Sutskever between labs illustrates how individual contributors can shift competitive balance This is why hiring and retention remain central to strategy For observers the takeaway is that the AGI race has been shaped by personal relationships and competitive fear as much as by technical progress For labs the lesson is that talent and infrastructure remain the primary bottlenecks Do you think the focus on slowing competitors is a rational safety strategy or does it undermine open research and progress Share your view in the comments
By Behind the Tech6 months ago in Futurism
Career Expert Says Soft Skills Are Now the Main Filter for Hiring as AI Takes Over Technical Tasks
Read Time 6 minutes Tags Interviews Career Advice Soft Skills AI Hiring Behavioral Questions STAR Method In today’s job market having a positive collaborative attitude is just as important as having a polished resume says career expert Erin McGoff Soft skills are top of mind for hiring teams according to McGoff the author of The Secret Language of Work Hyper Helpful Scripts for Every Situation That is why hiring managers often ask behavioral questions like Tell me about a time you disagreed with a boss or coworker in order to learn more about how you react to different situations As we move into an age where AI can take on more technical skills interpersonal skills are something that cannot be replaced McGoff says Companies are really prioritizing attitude personality culture fit because other things can be taught Your goal is to demonstrate your capacity to maturely navigate workplace conflict How to answer the disagreement question One Keep it professional not personal McGoff number one tip for answering this question is to keep it professional Instead of making it about personal differences you want to keep your response oriented towards the business she says it is the mature thing to do For example do not spend the entire conversation complaining about a past boss who denied your PTO request She also recommends framing your scenario as a difference in opinion rather than as an argument or dispute A candidate could start their answer by saying I have worked with many great bosses so while I have not had many personal disagreements there have definitely been instances where I have had to professionally advocate for alternative perspectives and ideas This framing shows judgment and avoids sounding like a complainer Two Use the STAR format to structure the story Your answer should center on a concrete example of a time you calmly and constructively resolved a professional disagreement McGoff recommends structuring the rest of your answer using the STAR format which stands for situation task action and result To describe the situation and task a candidate could say In my previous role there was a scenario where we were working on a project for a client The project was moving in a certain direction but I had specific insight that made me believe that a different direction would be more advantageous for this client Highlight the action you took to communicate your opinion and work toward a solution I asked my boss for a one on one and expressed this alternative path to them and made the case for why I thought it was better for the client Finally share how you solved the issue and emphasize the positive result We decided to compromise on the approach and move forward The client was really happy with the results and the project was a huge success You can also share what you learned from the experience or how you adapted your approach or workflow to avoid similar issues going forward Three Show healthy conflict not ego The point of your anecdote is not to show that you proved your boss wrong Instead it is an opportunity to demonstrate your ability to handle healthy conflict Healthy conflict is how we get work done she says You have to learn how to disagree with people in a professional way or else you will never rise up in your career Why this matters more now One AI shifts hiring criteria As AI takes over routine technical tasks companies are re weighting what they look for in candidates Technical skills can be taught or augmented with tools but judgment communication and conflict resolution are harder to automate Hiring teams use behavioral questions to test for these traits because they predict how a candidate will operate under stress and ambiguity Two Risk of red flag answers McGoff warns that answers focused on personal grievances or drama are red flags They signal low emotional regulation and poor fit for collaborative environments In a market where many candidates have similar technical backgrounds the differentiator is how you handle disagreement and ambiguity Three Transferability across roles The ability to advocate for an alternative view without damaging relationships is valuable in engineering product design sales and management It shows you can influence without authority and that you prioritize business outcomes over ego Practical preparation steps One Prepare two stories in advance Have one story about a disagreement with a manager and one about a disagreement with a peer Make sure both have clear business context and a positive outcome Avoid stories involving confidential information or ongoing legal issues Two Practice brevity and structure Interviews move fast A STAR response should take 60 to 90 seconds Start with one sentence for situation and task one or two sentences for action and one sentence for result End with a one sentence takeaway about what you learned Three Align with company values Research the company culture before the interview If the company values customer obsession use a story where you advocated for the client If it values speed and ownership use a story where you unblocked a project by resolving a disagreement Common mistakes to avoid One Making it personal Complaining about personality clashes signals risk The interviewer is asking about professional judgment not workplace gossip Two No resolution Stories that end in stalemate or escalation suggest poor conflict management Hiring managers want to see you can reach compromise or influence decision makers Three Over claiming credit Claiming you single handedly changed the outcome sounds unrealistic Focus on your role in the resolution and give credit to collaboration where due What hiring managers are really testing One Judgment Can you distinguish between issues worth escalating and issues to let go Two Communication Can you explain your view clearly and listen to counterarguments Three Accountability Do you take responsibility for outcomes and learn from the process For candidates the shift means investing in communication training feedback skills and stakeholder management is now as important as learning a new framework or programming language For employers it means updating interview rubrics to score behavioral responses consistently rather than relying on gut feel Do you think AI will make behavioral interviews even more important in the next two years or will new tools change how hiring works Share your view in the comments
By Behind the Tech6 months ago in Futurism
US Proposes AI Safety Protocol With China Citing Lead in Frontier Model Development
Read Time 6 minutes Tags AI Geopolitics US China AI Safety Export Controls Nvidia Anthropic US Treasury Secretary Scott Bessent told CNBC on May 14 2026 that Washington can hold discussions with Beijing on artificial intelligence because we are in the lead Bessent spoke from Beijing during President Donald Trump two day meeting with Chinese President Xi Jinping and said the two countries would establish a protocol for AI safety adding that they were having a wholesome dialogue because the US remains ahead in the race to develop frontier models The statement marks the first public US China engagement on AI risk at the presidential level in 2026 and frames the discussion around preventing non state actors from accessing advanced models What was announced One Protocol framework for AI safety Bessent said the two AI superpowers are gonna start talking We re gonna set up a protocol in terms of how do we go forward with best practices for AI to make sure non state actors don t get a hold of these models The scope was not detailed but the language aligns with prior discussions on model security access controls red teaming and incident sharing The reason we are able to have wholesome discussions with the Chinese on AI is because we are in the lead he added I do not think we would be having the same discussions if they were this far ahead of us This positions US export controls and current model advantage as leverage for negotiating norms Two Chip export policy remains unsettled Bessent was asked about a Reuters report that Washington had cleared sales of Nvidia H200 AI chips to several major Chinese technology firms He said there had been a lot of back and forth on the matter Nvidia CEO Jensen Huang joined Trump delegation to China as a late addition indicating that semiconductor supply is part of the negotiation US policy has restricted sales of advanced semiconductors to China since 2022 with periodic adjustments based on diplomatic and commercial calculations Any clearance for H200 would be significant for Chinese training capacity but politically sensitive in Washington Three Taiwan and broader security context Beijing readout said Xi emphasized that Taiwan is the most important issue for bilateral relations and warned against mishandling the issue Bessent told CNBC that Trump would say more on Taiwan in the coming days The AI talks are happening alongside unresolved trade and security disputes which limits the scope of any agreement to narrow technical issues Context on the US China AI balance One US lead in frontier capability Bessent claim of lead reflects current benchmark results where US labs OpenAI Google DeepMind Anthropic maintain advantage on general capability and reasoning Anthropic recent Mythos AI model has alarmed Washington due to reported cyberattack capabilities and the company decision to release it only to select business partners The US lead depends on access to advanced chips talent and compute infrastructure If export controls remain strict that gap may persist in the near term Two China response and domestic push China has prioritized domestic chip development and model training to reduce dependence on US technology Firms like Baidu Alibaba DeepSeek and ByteDance have released models with improving performance The gap is narrowing in some areas but restrictions on high end GPUs constrain training at the frontier scale Three Multilateral governance efforts The bilateral protocol fits into a broader pattern of states seeking norms for AI safety without slowing deployment The UK AI Safety Summit 2023 and UN AI advisory body work have pushed for risk based frameworks but enforcement remains limited A US China protocol could set a floor for model security and information sharing Implications for industry and policy One Impact on model releases Bessent said he anticipates a big step function jump in upcoming large language model releases from Google Gemini and OpenAI If true this would widen the capability gap and strengthen US negotiating position It also raises stakes for safety testing and responsible deployment Two Corporate exposure Nvidia and other chip firms remain exposed to policy shifts Any change in export rules affects revenue and supply chains The presence of Jensen Huang in Beijing signals that industry is seeking clarity on what is permissible Three Risk of leakage and misuse The stated goal of the protocol is to prevent non state actors from accessing advanced models This points to shared concern about model weights leaking and being used for cyberattacks or bioweapon assistance Both sides have an interest in avoiding catastrophic misuse even if they compete on commercial deployment What to watch next One Concrete measures The protocol needs to specify what counts as best practice Is it secure model storage mandatory red teaming watermarking for synthetic content or restrictions on fine tuning The absence of detail means the announcement is a starting point not a binding agreement Two Verification and trust Any agreement requires a way to verify compliance US and China have limited trust on technology issues so monitoring will be a sticking point One likely path is third party audits and information sharing on incidents rather than direct inspection Three Link to export controls If the US sees the protocol as working it may keep current chip restrictions in place If talks stall restrictions could tighten The connection between safety dialogue and trade policy makes this a high stakes negotiation For companies building frontier models the message is that safety is becoming a diplomatic issue not just a product feature For governments the test is whether a narrow technical agreement can survive wider strategic friction Do you think a US China AI safety protocol can work without addressing export controls on advanced chips Share your view in the comments
By Behind the Tech6 months ago in Futurism
Anthropic Launches Claude for Small Business to Push AI Adoption Beyond Fortune 500
Read Time 6 minutes Tags AI SMBs Anthropic Claude Cowork SaaS AI Adoption Business Automation Anthropic is looking to court smaller companies To that end the company announced on May 13 2026 the launch of Claude for Small Business a new suite of services designed for customers who less resemble Walmart and Starbucks and more resemble the local hardware store or coffee shop So far much of the most intensive AI adoption has occurred at the enterprise level In the recent past studies have shown that most companies that scaled AI systems beyond experimental or pilot level integration tended to be large companies with expansive budgets This appears to be changing somewhat as smaller and midsized businesses are seeing greater adoption Product details and positioning One Claude for Small Business feature set The new bundle of features is available via a newly introduced toggle within Claude Cowork the company task automation platform for business users that can browse the web manage files and execute multistep workflows on a user behalf By toggling it on paying users gain access to automated services including bookkeeping functions business insights and generative tools for ad campaigns The suite also includes integrations between Claude Cowork and software products like QuickBooks Canva Docusign HubSpot and PayPal These integrations let small businesses automate tasks that previously required manual entry or separate contractors Small businesses account for 44 percent of US GDP and employ nearly half the private sector workforce but their adoption of AI has lagged behind larger enterprises Anthropic said Tools and training are rarely tailored to the ways small businesses operate and as a result their use often stops at the chat window Two Go to market strategy Anthropic is planning to aggressively promote its new features with a coast to coast promotional tour starting in Chicago and hitting 10 cities in total At each stop the company plans to offer a free AI training workshop that will be available to 100 local small business leaders The goal is to move adoption beyond the chat window by showing owners how to automate real workflows Three Competitive context Anthropic is a little behind its competitor OpenAI which launched Enterprise ChatGPT at the end of 2023 including an integration for smaller teams called ChatGPT Business The move signals that the AI platform wars are expanding downmarket and that the next major battleground for user acquisition is not the Fortune 500 but the 36 million small businesses that make up the backbone of the US economy Why small business adoption has lagged One Cost and complexity Enterprise AI rollouts often require IT teams custom integrations and dedicated budgets Most small businesses do not have an IT department and cannot justify six figure contracts for AI tooling Two Product market fit gap Many early AI products were built for analysts engineers and marketers in large firms They assume structured data and repeatable processes Small businesses run on ad hoc workflows paper records and owner operator decision making A chat window alone does not solve that Three Training and trust Owners need to see concrete ROI in hours saved or revenue gained before they adopt Tools that can bookkeep reconcile invoices and draft local ad campaigns lower the barrier because the value is immediate and measurable Technical and product implications One Agent based workflows Claude Cowork uses agents that can browse the web manage files and execute multistep workflows on a user behalf This matters for small businesses because it reduces the need to switch between QuickBooks Canva and email to complete one task The agent acts as a glue layer Two Integration strategy By connecting to QuickBooks Canva Docusign HubSpot and PayPal Anthropic is meeting small businesses where they already work The integrations reduce friction and make the AI output actionable rather than just informational Three Pricing and distribution Anthropic has not disclosed pricing for the small business tier but the toggle is available to paying users of Claude Cowork The free workshops in 10 cities serve as both customer acquisition and training reducing the activation barrier that has slowed SMB adoption Market implications One Expansion of AI TAM Enterprise AI is a multi billion dollar market but small business AI could be larger by user count If even 10 percent of the 36 million US small businesses pay 50 dollars per month that is a 2 billion dollar annual revenue pool Two Pressure on vertical SaaS Companies like QuickBooks and HubSpot already offer AI features Anthropic entry turns them into both partners and competitors The outcome will depend on whether Claude Cowork becomes a hub that orchestrates these tools or just another chat interface Three Data and privacy considerations Small businesses hold sensitive financial and customer data Any AI platform handling bookkeeping and payroll data needs clear policies on data retention model training and access controls Anthropic has emphasized privacy in enterprise contexts but SMB adoption will test those policies at scale What to watch next One Conversion from workshop to paid use The 10 city tour will show whether hands on training converts owners into paying users Low activation has been the main reason SMB AI tools fail Two Feature depth Bookkeeping and ad generation are good starting points but stickiness will depend on whether Claude Cowork can handle invoicing inventory management and customer follow up without constant human oversight Three Competitive response OpenAI Microsoft and Google will likely respond with SMB focused bundles and deeper integrations in their own productivity suites The battleground is now the daily operating system of small businesses For Anthropic this is a bet that agents can do for small business what spreadsheets did in the 1980s For small business owners it is a chance to automate tasks that previously required hiring or late nights For the AI industry it marks the shift from pilot projects in large firms to real operational use in the broader economy Do you think AI agents can replace entry level admin work in small businesses without creating new compliance risks Share your view in the comments
By Behind the Tech6 months ago in Futurism
Recursive Superintelligence Startup Raises 650 Million Dollars to Automate AI Self Improvement
Read Time 6 minutes Tags AI Research Recursive Self Improvement AI Startups Autonomous AI Safety Notable researchers from Google Meta and OpenAI have launched a six month old company called Recursive Superintelligence to pursue the goal of building AI systems that can improve themselves with little or no human help The company has raised more than 650 million dollars from venture capital firms including GV Greycroft Nvidia and AMD and is now valued at more than 4 billion dollars despite having fewer than 30 employees The founders include Richard Socher who is also chief executive of You com and previously head of AI research at Salesforce and co founders Josh Tobin Jeff Clune and Tim Shi from OpenAI and Yuandong Tian from Meta The team also hired Peter Norvig who spent 25 years as director of research at Google and co wrote the standard AI textbook Artificial Intelligence A Modern Approach What recursive self improvement means One Definition and technical basis Recursion in computer science refers to a function that feeds itself After a recursive procedure generates information it uses that information to generate something else and so on In AI this means an AI system writes code that improves its own architecture training process or data pipeline without direct human engineering A veteran researcher Richard Socher said AI is code And now AI can code The ingredients are there The idea is to push more and more work onto machines including the generation of new ideas that drive AI development forward Two Current state of capability Companies like Anthropic and OpenAI released new AI systems late last year that were particularly good at writing computer code In recent months the technology has rapidly remade the way Silicon Valley engineers build test and modify new software applications If an artificial intelligence system can write code it can help accelerate the development of things as varied as word processors and social media apps OpenAI has said it is now building an automated AI researcher By the fall the company hopes to have a system that can do the work of a less experienced researcher said Sam Altman Similar efforts are underway at other leading companies Three Research focus on open endedness Many of the Recursive Superintelligence founders specialize in a kind of AI development called open endedness This involves building software systems that can run for days months or even years in pursuit of goals set by the researchers The aim is to create systems that explore large solution spaces without constant human guidance Company structure and challenges One Talent and funding density Recursive Superintelligence has fewer than 30 employees but is valued at more than 4 billion dollars The high valuation reflects both the seniority of the founders and investor belief that automating AI research could unlock compounding returns in capability The company operates offices in San Francisco and London Two Timeline and scope Dr Socher said his start up would need years to build the kind of technology that he and his co founders envisioned The company hopes to eventually apply the technology to other fields such as drug discovery and other kinds of biological research The founders acknowledge that current technology is a long way from the point where humans can be removed from the loop Humans must still generate the new ideas that drive AI development forward Three Competitive landscape Recursive Superintelligence is not alone in this pursuit The company should not be confused with Ricursive Intelligence which is pursuing a similar goal and is also valued at 4 billion dollars The prominent AI start ups Anthropic and OpenAI are also chasing recursive self improvement which has been an obsession among Silicon Valley technologists for decades Technical and safety considerations One Potential for capability acceleration If self improvement loops work even partially they could accelerate AI capability faster than linear scaling of compute and data This would shorten timelines to advanced systems and increase pressure on evaluation and safety processes Two Alignment and control risk Automated AI research raises the question of how to ensure that self modifying systems remain aligned with human intent A system that can rewrite its own training objective or architecture introduces new failure modes that are hard to audit or contain Three Evaluation bottleneck Current evaluation methods rely on human designed benchmarks and red teaming If AI begins to generate its own research agenda it may outpace human ability to assess risk This is why many researchers argue that interpretability and monitoring infrastructure must advance in parallel with capability Market and industry implications One Talent consolidation The involvement of researchers from OpenAI Meta and Google shows continued movement of senior talent into well funded startups focused on frontier research This concentrates expertise but also accelerates knowledge transfer outside large labs Two Investment thesis Investors are betting that the first group to automate AI research will gain a compounding advantage in both capability and cost The 650 million dollar raise signals confidence that the problem is now tractable with current models and tooling Three Application expansion The founders plan to apply the technology beyond code generation to fields like drug discovery and biological research If the approach works it could shorten development cycles in domains where experimentation is expensive and slow What to watch next One Benchmarks for autonomous research The field needs clear benchmarks for what counts as an autonomous AI researcher OpenAI target of a system that can do the work of a less experienced researcher by fall 2026 will be a test case Two Safety protocols As capability increases the need for secure model storage controlled access and robust monitoring becomes more urgent Any recursive system that modifies itself must have guardrails that cannot be easily removed by the system itself Three Regulatory response Governments are watching this area closely because recursive self improvement could change the pace of technological change faster than existing policy cycles The US China protocol discussed earlier this week is an early signal that states see AI autonomy as a diplomatic issue For researchers the challenge is to prove that recursive loops can produce real gains without losing control For investors the bet is that this is the next step after code generation models For the rest of the field it raises the question of how fast capability can move once humans are partially out of the loop Do you think recursive self improvement is closer than most people expect and should safety research get more funding than capability right now Share your view in the comments
By Behind the Tech6 months ago in Futurism
AI Did Not Crack Bitcoin But Helped User Recover 400000 Dollars By Finding Old Wallet File
Read Time 6 minutes Tags Bitcoin Crypto Recovery AI Claude Wallet Security Data Forensics A Bitcoin holder known as Cprkrn on X went viral on May 13 2026 after claiming he recovered around 400000 dollars in BTC from a wallet locked for more than a decade with help from Anthropic AI chatbot Claude The user said Claude helped him regain access to 5 Bitcoin after years of failed attempts At the time of writing Bitcoin was trading at around 79600 dollars putting the 5 BTC recovery at roughly 398000 dollars Holy fcking sht omg Claude just cracked this sht he wrote in the viral post thanking Anthropic and CEO Dario Amodei He even said he plans on naming his child after Amodei What actually happened One Lost password and old files The wallet had been locked since the user college days He said he originally bought the crypto when it was worth around 250 dollars per coin before losing access after changing the wallet password while high The password as later revealed was lol420fuckthePOLICE The user said he had tried for years to recover the funds claiming he ran through trillions of possible password combinations Two Claude role was data forensics not brute force The breakthrough came when he uploaded files from his old college computer into Claude Rather than simply guessing the password Claude helped dig through the old files and identify an older wallet dat file that appeared to predate the password change The user also reportedly had an old mnemonic phrase which helped unlock the wallet once the correct file was found I tried like 7 trillion passwords lmfao he wrote Found this old pneumonic a few weeks ago that ended up being the old password before I changed it Thought I was screwed Last ditch effort dumped my whole college computer into Claude It found an OLD wallet file that the pneumonic Three Tool context BTCRecover is a known wallet recovery tool designed for cases where users already know most of a wallet password or seed but need help testing variations Its documentation says it supports Bitcoin Core wallet recovery among several other wallet types Claude contribution was to locate the correct wallet dat file among a large set of old files and connect it to the mnemonic the user had found Why this is not a break of Bitcoin security One Cryptography remains intact Claude did not break Bitcoin elliptic curve cryptography or bypass the wallet encryption The system only works if you have the correct private key or seed phrase In this case the user already had the seed phrase from an earlier version of the wallet The problem was finding which wallet dat file matched that seed phrase Two Value of context aware search Large language models excel at parsing unstructured data across thousands of files and matching patterns A human would take days to manually inspect an old college hard drive for wallet files wallet backups and text fragments that look like seed phrases Claude automated that process by reading the files understanding file types and surfacing the relevant one Three Limits and risk This only works if the user has some existing credential material or an older backup If the wallet was truly lost with no seed phrase and no backup file AI cannot recover it The story spread quickly because it looks like AI hacking but it is closer to advanced data recovery and organization Broader implications for crypto and AI One Rise of AI assisted recovery Services that combine AI with wallet recovery tools are likely to grow Many users lose access due to forgotten passwords or misplaced backups not due to lost keys AI can reduce the time and cost of sifting through old drives email archives and cloud backups to find fragments of credential data Two Security practices need updating The case highlights why users should store backups in multiple secure locations and avoid changing passwords in an impaired state It also shows why old hard drives are valuable targets If AI can parse them quickly so can attackers Three Responsible disclosure Anthropic has not commented publicly on the case but the incident will raise questions about how much file access AI assistants should have and whether uploading entire drives creates privacy risk Users need to understand that sending data to a cloud AI means that data leaves their device What this means for users with lost crypto One Check for old backups Before trying brute force methods search for wallet dat files keystore files txt files with seed phrases and old email exports AI can help but only if the material exists Two Use dedicated tools BTCRecover and similar tools are designed for password variation testing If you know part of the password or seed they can automate the combinatorics faster than manual methods Three Secure recovered funds Once access is restored move funds to a new wallet with a secure backup process Hardware wallets and offline storage reduce future risk The story resonated because it mixes crypto nostalgia AI capability and a happy ending The reality is more mundane AI did not break math it found a file The distinction matters because it sets realistic expectations for what AI can and cannot do in crypto recovery Do you think AI assistants should be allowed to scan entire local drives for recovery purposes or is the privacy risk too high Share your view in the comments
By Behind the Tech6 months ago in Futurism











