It’s almost hard to remember, but just a few years ago, a lot of very smart people thought that science was in terminal decline. Respected researchers were writing papers with titles like “Are Ideas Getting Harder to Find?”, “Papers and patents are becoming less disruptive over time”, and “Slowed canonical progress in large fields of […]

The post Love it or loathe it, AI is birthing a golden age of science appeared first on Asia Times.

• A former Anthropic employee wrote a Substack post about his 8-month stint at the then-new AI lab.
• It offers a rare look at Anthropic's early, safety-obsessed culture that the author said left him burned out.
• He describes a visit from the FBI and fresh croissants in an ugly office.

Anthropic's executives raced to build a safe AI — and one former employee said that pursuit left him feeling burned out and disillusioned.

In a lengthy Substack post, Ivan Vendrov, who worked as a member of the AI lab's technical staff from late 2022 to mid-2023, described an intense workplace, where fears of AI catastrophe clashed with the urge to build safe AGI for commercial use.

Vendrov began the post by saying his account should not be taken as gospel. He said he recalled the events "from memory," wrote the "gist of the dialogue rather than the exact words used," and that the post is "merely the experience of one guy in a junior role during a short window of time."

Still, it's a remarkable look into the early days of one of the world's preeminent AI companies, with a named employee explaining how the company's formative period created a highly particular workplace.

It also comes days after Anthropic filed to go public. Anthropic declined to comment on this story.

Equity, the FBI, and a sushi dinner

Vendrov, whose LinkedIn says he now works at Midjourney, wrote that he was weighing contract offers from OpenAI and Anthropic in 2022.

He says he decided to join Anthropic after lunch with one of the AI company's cofounders, who said they were already chatting with regulators in Washington.

A recruiter asked whether he wanted to pledge two-thirds of his equity to charity.

"All the co-founders have made the pledge," Vendrov recalled the recruiter saying. He wrote that he was unsure whether the job offer was contingent on his agreement, so he said yes.

Anthropic's S-1 filing describes an employee donation-matching program, where workers can pledge a "meaningful portion" of their equity awards. The company reported that 50% of its employees are included in the program.

Vendrov described a cramped, ugly office. He said it was stocked with fresh croissants and, later, a popular masseuse came twice a week. He said he was never able to schedule a massage because it was always fully booked.

There was a visit from the FBI once, he wrote. The agent warned that Anthropic was "probably already the target of multiple foreign intelligence agencies."

Anthropic also hosted a sushi chef in the office, Vendrov wrote — but the luxe experience was met with resistance. The following day, he said a cofounder wrote in the company Slack that it set the wrong tone.

"We're not here to get rich and eat caviar," they said, per Vendrov's memory, before pitching future dinners of rice and beans.

The burnout

Vendrov's post detailed sleepless nights, walks with other stressed-out colleagues, and moments where he felt he couldn't be honest about his emotions with colleagues.

Eight months after signing his contract offer, Vendrov said he wrote a resignation letter. As he drafted the letter, he said he thought about describing his doubts to the company, but worried it would create doubt in colleagues who were "smarter than me, more competent than me, more solidly well-intentioned than me."

Instead, he said he praised Anthropic as an "unbelievably good organization."

Vendrov did not respond to a request for comment from Business Insider.

Um estudo interno divulgado pela Anthropic revelou que os agentes de inteligência artificial da empresa já trabalham seis horas para cada hora de um funcionário humano. O resultado direto desse avanço é queda no número de contratações.

Os agentes são ferramentas capazes de pensar e tomar decisões de forma autônoma, com pouquíssima interferência humana.

O especialista em IA e colunista da CNN Brasil Álvaro Machado Dias explicou que, nesse novo cenário, o papel do trabalhador humano se aproxima cada vez mais da gestão. “Ele faz o trabalho que era do chefe dele”, afirmou Machado.

Segundo ele, as principais funções humanas passam a ser escrever planos para que a IA execute tarefas, checar os resultados produzidos e avaliar se é possível avançar para a próxima etapa ou se são necessários ajustes no caminho.

Mudança no perfil das contratações

As empresas estão contratando menos, mas buscam profissionais com expertise técnica capaz de ser aplicada diretamente em fluxos de inteligência artificial.

“Há uma mudança no perfil das contratações e isso está mexendo com todos os departamentos”, destacou Machado.

Para ele, departamentos inteiros, como o de marketing, já não existem da mesma forma que há dois anos. Nesse contexto, profissionais iniciantes perdem espaço, pois as empresas passam a priorizar quem já chega com habilidades técnicas prontas para serem aplicadas.

Machado apontou ainda que a queda de 9% nas contratações observada no setor de tecnologia tende a se expandir para o mercado como um todo, especialmente nas economias mais avançadas.

“Mas uma hora, com certeza, chega aqui”, alertou, referindo-se ao Brasil. No país, 50% das empresas já utilizam inteligência artificial, e a perspectiva é de crescimento acelerado.

IA avança além do setor de tecnologia

Segundo o especialista, a adoção de agentes de IA em setores fora da tecnologia pode ser ainda mais fácil de implementar do que nas próprias empresas desenvolvedoras. Isso porque, quanto mais tarefas repetitivas um setor possui, maior é o potencial de automação.

Ele citou exemplos práticos: em bancos, agentes podem realizar análises de crédito verificando documentos automaticamente; em escritórios de contabilidade, podem preencher declarações de imposto de renda; e em lojas, podem atender clientes via WhatsApp e checar o inventário simultaneamente.

O ser humano como gargalo

Um dos pontos mais relevantes do debate, que aconteceu durante o Prime Time, da CNN Brasil, foi a confiança nas máquinas.

Se antes a desconfiança vinha dos erros frequentes da IA, agora o risco surge do excesso de confiança. “A desconfiança emerge porque as pessoas estão simplesmente mandando trabalho para o mundo sem checar”, afirmou Machado.

Ele destacou que a frase que circula no Vale do Silício atualmente é que “o ser humano se tornou o novo gargalo”, já que a IA gera análises, relatórios e insights em volume e velocidade que superam a capacidade humana de revisão.

Quanto ao perfil ideal do profissional nesse novo ambiente, o especialista descreveu alguém com muita velocidade e capacidade de execução, mas que não seja excessivamente detalhista. “A tendência atual é as coisas saírem com menos checagem e maior volume”, disse.

Para ele, isso implica mais incerteza e mais risco, mas, no agregado, a produtividade é multiplicada de forma significativa.

49% dos brasileiros acham que IA melhora desempenho no trabalho

• Mathematicians need to develop their oft-neglected social skills, a math professor at Cornell said.
• Though AI can do the legwork on complex problems, personal charisma is a human-only skill, Steven Strogatz said.
• "Rizz is more and more in demand," he added.

"Social skills" and "mathematics" are terms not often grouped together. In the age of AI, one math scholar says that needs to change.

With AI now capable of handling the geeky legwork required by complex problems, Steven Strogatz, a lecturer and Susan and Barton Winokur Distinguished Professor at Cornell University, said mathematicians should prioritize developing their personal charm in order to set themselves apart.

"I'm starting to think that what makes a future mathematician or a math professor is your unique brand," Strogatz said in an interview with Grant Sanderson, a math educator affiliated with Stanford University.

"Rizz is in more and more demand because it's something that only people have," he added.

Rizz, a Gen Z slang term, is defined by Merriam-Webster as "romantic appeal or charm." (Strogatz, who is 67, appeared to use the term solely to mean charm or charisma. He did not suggest mathematicians need to work on their romantic prowess.)

"What we're selecting for is an element of mathematical rizz," said Sanderson. "The PhDs to be are not going to be differentiated by pure problem-solving abilities."

The brusque push of AI companies into mathematics has caused consternation and head-scratching from many a math geek, as the field ponders ways to stay relevant while OpenAI floods the zone with hundreds of proposed solutions to the world's most complex problems.

Some professors have sworn off AI altogether. Most, however, are trying to find ways to incorporate the technology into their work.

In September, a group of undergraduates at Caltech announced a "mathathon" event in which participants were encouraged to use AI to solve problems in a single, no-sleep sprint. Only days after the undergrads announced the event, a coalition of over 700 academics signed an open letter calling for its cancellation, citing its promotion of AI "slop."

Though the event is set to continue, with some slight alterations, the clash underscored an internal conflict in the math community over how to adapt to AI — a development poised to impact many burgeoning careers in the field.

"We've had whole research programs wiped out on Tuesday," Tristan Buckmaster, a math professor at NYU, said after OpenAI published 700 new findings about outstanding problems earlier this month.

The issue is of supreme importance to Strogatz, who has made several media appearances in the last couple of months. In an essay published in The Atlantic on Sunday, he put the issue in stark terms.

"I am a math professor, and my field is in crisis. Nobody in math is ever going to forget the year 2026," he wrote. "Unless this is really the end of the subject, and then nobody will remember anything about math anyway."

La Tierra se mueve de tres maneras: sobre sí misma, con la rotación; alrededor del Sol, con la traslación; y oscilando como una peonza, en un lentísimo movimiento que tarda en completarse unos 26.000 años y que se conoce como precesión. Debido a su existencia, el cielo de hoy es distinto al que vieron otros hombres y mujeres a lo largo de la historia. Con el esoterismo europeo de la primera mitad del siglo XX se popularizó la idea de que las eras astrológicas marcaban el signo de los tiempos, según donde estuviera el Sol en el primer día de primavera: la Antigua Grecia y la República romana coincidieron con la era de Aries, el cristianismo con la de Piscis y pronto llegará la de Acuario, si no lo ha hecho ya. La New Age se subió al carro, y Raphael se desató cantando aquello de Aquarius. Los 2.000 años de duración de cada era son flexibles: Carl Jung, por ejemplo, situaba el inicio de Acuario entre 2000 y 2200. La astrología tradicional, que prefiere medir los cambios de época con las conjunciones de los señores del tiempo, Júpiter y Saturno, que se unen ópticamente en el cielo cada 20 años, dice que el momento importante fue el 21 de diciembre de 2020, cuando se produjo la primera “gran conjunción” en Acuario desde la Edad Media. Las redes pensaron que, esta vez sí, iba a comenzar el nuevo tiempo, pero muy alejado de la paz y el amor que profetizó la interpretación hippie. Simbólicamente, Acuario es la ciencia y la tecnología, el progreso, la inteligencia, la comunidad y la revolución, pero todo ello de una forma desapegada y etérea, incluso implacable, porque es un signo de aire. El wifi de casa, por ejemplo, sería muy Acuario.

Seguir leyendo

A divulgação do prospeto S-1, documento que antecede a entrada em bolsa, pela Anthropic, que detém o modelo Claude, na segunda-feira 28 de outubro, está a “reacender o entusiasmo” em torno da inteligência artificial (IA), afirma o analista da ActivTrades, Henrique Valente.

Este entusiasmo está refletido no avanço em 126 pontos do índice bolsista norte-americano Nasdaq na manhã desta quarta-feira 28 de outubro, adianta. “A publicação do prospeto S-1 da Anthropic, em preparação para a sua entrada em bolsa, ajudou a reacender o entusiasmo em torno da inteligência artificial, também alimentado por novos anúncios de financiamento e pelos resultados dos fabricantes de semicondutores”, referiu o analista.

No prospeto, a Anthropic revela que vai entrar em bolsa através do índice bolsista Nasdaq, com o símbolo ANT. Também foi divulgado que a empresa teve um prejuízo de 80,7 mil milhões de dólares (72,4 mil milhões de euros) nos primeiros seis meses do ano, quase o dobro face ao prejuízo de 41,9 mil milhões de dólares (37,6 mil milhões de euros) registado em 2025.

“A nossa receita foi de 16,3 mil milhões de dólares (14,6 mil milhões de euros) nos seis meses findos em 30 de junho de 2026, um aumento em relação aos 1,2 mil milhões de dólares (mil milhões de euros) registados nos seis meses findos em 30 de junho de 2025. O nosso prejuízo operacional foi de 693 milhões de dólares (622,3 milhões de euros) nos seis meses findos em 30 de junho de 2026, uma redução em relação aos 2,1 mil milhões de dólares (1,8 mil milhões de euros) registados nos seis meses findos em 30 de junho de 2025”, assinalou.

O documento mostra ainda que no final de setembro a empresa detentora do Claude tinha mais de 11.900 clientes com um volume de negócios igual ou superior a 100.000 dólares (89 mil euros) nos últimos doze meses, face aos mais de 950 do período homólogo. “Além disso, à medida que os clientes incorporam o Claude nos seus fluxos de trabalho mais críticos e iniciativas de crescimento, muitos expandem rapidamente as suas implementações para níveis de faturação mais elevados. Em 30 de setembro de 2026, tínhamos mais de 1.900 clientes com um volume de negócios igual ou superior a um milhão de dólares (900 mil euros) nos últimos doze meses e aproximadamente 250 clientes com um volume de negócios igual ou superior a 10 milhões de dólares (8,9 milhões de euros) nos últimos doze meses”, acrescentou.

A empresa explicou que este crescimento nos clientes é justificado pelo cada vez maior das soluções da Anthropic. “O valor que o Claude oferece aos clientes contribuiu para a nossa taxa de retenção de receitas líquidas de aproximadamente 900% em 30 de setembro de 2026”, assinalou. Foi ainda revelado que a receita aumentou 1.368% em termos homólogos nos três meses que acabaram em 30 de junho de 2026, enquanto que as despesas de computação e infraestruturas aumentaram 300% no mesmo período temporal.

O entusiasmo nos mercados de capital com a Anthropic espalhou-se a outras regiões. “Na Ásia, a SK Hynix e a Samsung Electronics chegaram a subir cerca de 8% esta manhã (28 de outubro). Na Europa, a ASML ganhou 2,8% na sessão de ontem (27 de outubro). Nos Estados Unidos, a Broadcom avançou 4% e a CoreWeave valorizou 7,9%, enquanto o Nasdaq Composite terminou a sessão [de 27 de outubro] com uma subida de aproximadamente 1,2%”, descreveu o analista da ActivTrades Europe.

Agora o foco está na reunião da Reserva Federal norte-americana (Fed), marcada para esta quarta-feira 28 de outubro, onde se decidirá o caminho das taxas de juro, acrescentou. “A expectativa dominante é de manutenção das taxas de juro no intervalo de 3,75% a 4,00%, após a subida de setembro. Sem atualização do dot plot nesta reunião, o foco estará no comunicado e na conferência de imprensa de Kevin Warsh, às 18h30. Perante a relutância do presidente da Fed em dar indicações explícitas sobre a trajetória futura das taxas de juro, os investidores procurarão sobretudo perceber se a pausa poderá prolongar-se ou se a persistência da inflação deixa aberta a porta a uma nova subida em dezembro”, conclui Henrique Valente.

• Rishi Sunak has senior advisor roles with Anthropic, Microsoft, and Goldman Sachs.
• Sunak, who was the UK's prime minister between 2022 and 2024, makes up to $10,000 an hour from them.
• Goldman paid him about $1.5 million, compared with Anthropic shares valued at about $500,000.

How much is Anthropic paying for the advice of the UK's former prime minister? A lot less than Microsoft.

Rishi Sunak, who led the country from 2022 to 2024, last year took on senior advisor roles with Anthropic, Microsoft, and Goldman Sachs.

As he still serves as a member of Parliament, the payments are detailed in a public register of MPs' financial interests. The most recent register, published Monday, shows Sunak earned more than £2.2 million, or over $2.9 million, from the three jobs.

Sunak, who started his career as a Goldman analyst before completing an MBA at Stanford, donated these earnings to charity. He and his wife Akshata Murthy set up The Richmond Project, which works to improve people's numeracy skills.

The couple is jointly worth about $750 million, according to The Sunday Times Rich List, with much of that wealth coming from Murthy, whose father founded Indian tech firm Infosys.

Anthropic hired Sunak as a senior advisor last October, paying him with equity in the AI firm. The register shows the shares were worth £373,000, or $495,000, when his one-year contract began.

Working an estimated 70 hours a year, that translates to about $7,000 an hour.

Also last October, Sunak started advising Microsoft. The tech giant paid him £741,735 a year, or almost $1 million. For an estimated 96 hours of work, per the register, that works out to about $10,200 an hour.

Sunak's role with Goldman is the most lucrative. He was paid nearly £1.1 million a year, or about $1.5 million.

The register shows the role involves an estimated 175 hours of work a year, suggesting he earns about $8,300 an hour.

Alongside his work in tech, Sunak writes a column for The Sunday Times. He also donates this money to charity, but it pales in comparison to the income from his other jobs. The register shows he earns £28,500 a quarter for five hours of work a week — about $580 an hour.

Sunak's office did not immediately reply to a request for comment.

• An Anthropic manager says Claude has made traditional coding interviews a poor hiring signal.
• His advice to juniors: target business problems and build a focused prototype over a weekend.
• Google and Amazon are also reshaping engineering work and hiring around AI fluency and impact.

Out with the coding test, in with Claude.

That's the shift Felix Rieseberg, an Anthropic engineering manager, says he has made to his team's hiring process, arguing that AI has rendered traditional coding interviews a poor measure of who will succeed at work.

"The coding interview is dead, at least for my team at Anthropic," Rieseberg wrote in a Tuesday post on X. "I've seen enough to know that it no longer provides signal."

Rieseberg elaborated in a blog post, writing that there's "no point" in giving candidates coding challenges or asking them about programming languages like assembly.

"Claude is now better at both," he said of the company's own AI model.

Coding challenges were once a "weak but useful" predictor of success, Rieseberg said, but the most impactful builders on his teams are not necessarily the people who excel at such tests.

"I'm seeing engineers who are comparatively new fly past people who are probably better coders and have been around much longer," he wrote. "Hunger beats tenure."

AI is writing most of the code at Anthropic, and the shift tracks with changes underway at other Big Tech companies. Business Insider's Hugh Langley reported last week that Google is expanding pilots for software-engineering interviews that allow candidates to use Gemini to assess their AI fluency and prompting skills.

Amazon, meanwhile, told engineers on Tuesday to prioritize customer and business outcomes over raw AI usage. The company calls the moment the era of the "Renaissance Developer," in which human judgment and machine intelligence combine to build products faster.

How junior engineers can stand out

The rise of AI coding tools has fueled fears that junior developer roles will become harder to get, as AI takes on work like debugging and building simple features — tasks that once helped early-career engineers learn the ropes.

In his blog, Rieseberg shared advice for junior engineers — or, more broadly, people without organizational status or power. Rather than just clearing backlog tickets, he said they should identify the business problems their company is trying to solve.

Then, they should come up with a bold idea that could materially advance those goals and build a working prototype over a weekend. The project should communicate one compelling idea, he said, rather than bury several good ones inside an overly ambitious product.

Newer engineers should also lean into their outsider perspective. They may be more likely to spot broken user experiences or question ingrained assumptions about what is possible, Rieseberg said.

He pointed to two relatively new Anthropic engineers who built a graphical interface for Claude Code over a weekend; the first public version shipped around eight days later.

Rieseberg said he has not yet solved the problem of how to assess candidates who build effectively with AI.

For now, he said he is scanning resumes and portfolios for small prototypes that show "taste, empathy, and principled thinking" — qualities that AI researchers and tech leaders are increasingly putting a premium on in the age of AI.

Poucas empresas de capital aberto afirmam ter o futuro da humanidade em jogo enquanto tomam decisões corriqueiras sobre a composição de seu conselho.

Não é o caso da Anthropic. Tendo alertado que sua inteligência artificial poderia um dia curar o câncer ou matar até 8 bilhões de pessoas, a Anthropic estabeleceu um plano complexo sobre quem poderá governar a empresa à medida que sua tecnologia avança rapidamente, conforme revelou seu pedido de IPO na segunda-feira.

A expectativa é que a startup seja avaliada em US$ 1,5 trilhão, o que a tornaria a empresa de benefício público mais valiosa do mundo. Segundo a legislação de Delaware, essa estrutura exige que o laboratório de IA equilibre os interesses dos investidores com a promoção de um bem maior.

Mais do que detalhes técnicos, a estrutura da Anthropic pode ter grandes implicações para investidores e para o público que adota cada vez mais a IA da empresa sediada em São Francisco. Os detalhes mostram como a desenvolvedora se isolou da pressão dos investidores para poder seguir o que considera ser o melhor para a humanidade.

A Anthropic concedeu 50,1% do poder de voto em determinadas questões aos seus sete cofundadores, que, segundo o documento, estão “distintamente capacitados para serem gestores da nossa missão”, conforme relatado anteriormente pela Reuters.

Embora um fundo independente controle a maioria das cadeiras do conselho, o poder de voto dos cofundadores se estenderá às eleições dos demais diretores e às propostas dos acionistas em geral.

O documento apresentado na segunda-feira revelou ainda mais complexidades na gestão de um negócio que a Anthropic espera que sobreviva à equipe unida que o criou.

Assim que a maioria dos cofundadores deixar o cargo e suas ações com direito a voto múltiplo “Classe F” forem vendidas, o conselho da Anthropic se dividirá em três grupos de tamanho semelhante, com mandatos de três anos.

Ao adotar um sistema de reeleição de diretores semelhante ao do Senado dos EUA, a Anthropic adotou uma estrutura que já foi popular como forma de impedir aquisições hostis. A empresa há muito expressa preocupação com os riscos de a IA cair em mãos autoritárias ou nefastas.

Essa estrutura também contraria algumas tendências de governança corporativa. Neste século, as principais empresas americanas passaram a eleger todos os seus diretores anualmente, se tornando mais receptivas às demandas dos acionistas em áreas como remuneração de executivos ou estratégia.

A Anthropic se recusou a comentar.

Flexibilidade a longo prazo

Além do seu conselho administrativo, a Anthropic possui outro órgão de governança com seu próprio conjunto de regras: seu fundo fiduciário de benefícios de longo prazo.

Atualmente presidido pelo líder global em saúde Neil “Buddy” Shah, o fundo controla uma classe especial de ações chamada “Classe T”, que lhe permite eleger a maioria do conselho da Anthropic. A desenvolvedora também já considerou o que aconteceria se essas ações deixassem de existir.

O objetivo de tais eventualidades é “preservar a flexibilidade a longo prazo caso as circunstâncias mudem”, afirmou a empresa em seu documento. Mesmo que alguém não vinculado à missão da Anthropic de beneficiar a humanidade adquira suas ações Classe T, a estrutura exigiria que os diretores seguissem “disposições relacionadas ao nosso propósito de benefício público”.

Em última análise, o poder do fundo sobre o conselho faz da Anthropic uma “empresa controlada” segundo as regras da Nasdaq, se isentando de cumprir alguns requisitos típicos, como ter um comitê independente de nomeação e governança corporativa. A Anthropic está usando essa isenção para colocar Daniela Amodei, presidente e irmã do CEO Dario Amodei, nesse comitê.

Os líderes no comando da Anthropic

A Anthropic também criou um comitê de segurança e risco em seu conselho, atualmente composto pelo CEO da Novartis, Vas Narasimhan, e pelo ex-funcionário da Casa Branca de Trump, Chris Liddell. O bilionário Reed Hastings, cofundador da Netflix, é o atual diretor independente principal da Anthropic.

Ainda assim, os sete cofundadores da Anthropic — incluindo os irmãos Amodei — permanecem no centro da governança num futuro próximo. Isso se deve, em parte, à existência de uma entidade chamada “Founder LLC”, que, depois do fundo fiduciário, terá a maior influência nas eleições do conselho.

Anthropic, the AI startup known for its Claude chatbot and its doomer CEO Dario Amodei, has filed for an IPO that could value the company at up to $2 trillion. The company's IPO filing devotes nearly 100 pages to the dangers its own technology could pose.

The post Hacking & Blackmail: 6 Risks Anthropic Flags in Its IPO Documents appeared first on Breitbart.

• Sam Bankman-Fried's Anthropic stake was liquidated in bankruptcy court two years ago.
• The feds seized shares belonging to two co-conspirators, Caroline Ellison and Nishad Singh.
• The government quietly sold those Anthropic shares last year. It's unclear where the money will go.

A little over a year ago, the federal government acquired a stake in Anthropic without paying a penny.

The shares were seized from associates of the crypto fraudster Sam Bankman-Fried. From there, they disappeared into a legal black hole.

As Anthropic rockets toward what could be the biggest IPO in history, those shares could now be worth billions of dollars.

Bankman-Fried's own Anthropic stake was liquidated in the bankruptcy of his failed cryptocurrency exchange. The Anthropic equity owned by Caroline Ellison and Nishad Singh, two associates who invested alongside him, took a different path. The government seized those shares and sold them to existing Anthropic shareholders, according to a person familiar with the sale.

It's still unclear whether the proceeds will be paid out to victims of the $11 billion fraud case, or if the feds will just keep the money.

It also remains a mystery which investors bought Singh's and Ellison's stock and how much they paid, but Anthropic shares have been on such a dizzying climb that it's likely those investors made a killing on them.

Bankman-Fried was convicted in 2023 on fraud and money laundering charges and sentenced to 25 years in prison after the collapse of his cryptocurrency empire. Prosecutors said he used his crypto hedge fund, Alameda Research, to siphon billions of dollars from customers of FTX, his crypto exchange.

The story of the crypto criminals' Anthropic shares is a tale with extraordinary elements, including one of the biggest financial frauds in modern history, a messy bankruptcy docket, and a company that experienced one of the fastest appreciations in business history as the world suddenly realized AI's existential importance.

An intelligent investment

Bankman-Fried, Singh, and Ellison each invested in Anthropic's 2022 Series B funding round. Bankman-Fried bought $500 million worth, which according to court records represented 13.56% of Anthropic at the time. Singh acquired $40 million in shares, and Ellison acquired $10 million in shares, court records reviewed by Business Insider show.

The artificial intelligence company's valuation has soared over the past four years. On the secondary market, Anthropic has been valued at $1.5 trillion. It is the most valuable private company in the world, according to Crunchbase.

The Anthropic shares purchased by Singh and Ellison could together be worth between $4.17 billion and $5.03 billion today, based on the $965 billion valuation the company announced this May, according to Olav Sorenson, who teaches venture capital strategy at UCLA. Harrison Rolfes, an analyst at PitchBook, put the combined figure at $2.62 billion. If Anthropic went public at a $2 trillion valuation, the shares would be worth about $5.44 billion, Rolfes said.

Four of Bankman-Fried's close friends and executives at his companies pleaded guilty as co-conspirators. Three of them, including Singh and Ellison, testified against him.

Ellison was the CEO of Alameda Research, which traded and invested funds that belonged to FTX depositors. She was also Bankman-Fried's on-and-off romantic partner. Singh, an FTX executive and early employee, misrepresented the exchange's revenue to investors.

After the collapse of FTX, Bankman-Fried's Anthropic shares — which were owned by Clifton Bay, an entity affiliated with Alameda Research — ended up in bankruptcy court along with the other dredged-up remains of his companies. His Anthropic shares were liquidated to pay FTX's creditors.

The estate in 2024 sold Bankman-Fried's Anthropic shares to a few dozen buyers for a total of $1.3 billion, more than double what he paid. The largest stake went to an entity affiliated with a United Arab Emirates sovereign wealth fund, bankruptcy court filings show.

The feds take a stake in Anthropic

As part of their sentences, a judge required Singh and Ellison to forfeit their Anthropic shares, which prosecutors said could be considered proceeds of their crimes.

At Singh's sentencing hearing, one of his attorneys, Andrew Goldstein, said Singh purchased his shares before participating in the criminal conspiracy and "he actually may have had a legitimate legal claim" to the shares but agreed to give them up as part of his plea agreement "because it was the right thing to do."

Reached for comment for this story, Goldstein told Business Insider that Singh hopes the government is able to quickly distribute proceeds of the sale to FTX victims. An attorney for Ellison declined to comment.

A federal judge ordered Ellison's and Singh's Anthropic shares to be transferred to the federal government, which took ownership of Ellison's shares in February 2025 and Singh's in April of that year, according to previously unreported court records.

Ordinarily, victims of crimes are compensated through a restitution process, which is overseen by courts. But the number of potential FTX victims could be in the millions, prosecutors said in court filings. As a result, the judge ruled, victim compensation would be handled through a process called remission, which is overseen by the Justice Department.

During Ellison's sentencing hearing, Justice Department prosecutors told the judge that the DOJ would either set up its own claims administration process to compensate victims, or work with the FTX bankruptcy process to identify victims and provide forfeited funds to them.

At the time, FTX's bankruptcy process was in full swing. A Delaware court appointed Sullivan & Cromwell, the elite Wall Street law firm, to untangle the company's assets, figure out who was owed money, and pay them.

The creditors in the FTX bankruptcy, prosecutors noted, largely overlapped with FTX's victims. They were generally depositors, lenders, and investors who were defrauded by Bankman-Fried and other executives. Prosecutors said the Justice Department could work with the bankruptcy estate's lawyers to get money back to them, as it had in previous large-scale financial frauds such as Bernie Madoff's Ponzi scheme.

There was an unusual twist that set the FTX bankruptcy apart. Bankman-Fried's investments — especially the Anthropic shares, along with some cryptocurrency — had grown substantially in value since FTX's collapse. Earlier in 2024, the Sullivan & Cromwell lawyers who'd taken over FTX projected that all the creditors would be repaid in full, with interest.

The Marshals take control

While prosecutors said they intended to use Singh's and Ellison's Anthropic shares for remission, the Justice Department could technically do whatever it wanted with them, according to Duncan Levin, a white-collar defense attorney who teaches a course on forfeiture at Harvard Law School.

"It's a very opaque process," he said. "It's completely at the discretion, by law, of the attorney general of the United States."

Typically, when the feds seize shares of private companies through criminal asset forfeiture, they send the shares to the US Marshals Service Complex Assets Unit for liquidation. The unit tries to value the shares as any other investor would, said Michael Bachner, a white-collar criminal and securities litigation attorney.

"They may look to: What would an institutional purchaser pay for these securities?" Bachner said. "Are there funds that are valuing the securities? Is there a secondary market already out there?"

For the sale of the Anthropic shares, timing was crucial. The Marshals Service is supposed to preserve as much value as possible, Bachner said. By the time the feds got hold of the shares, Anthropic had become an economy-shifting AI giant. In its March 2025 Series E fundraising round, Anthropic was worth $61.5 billion. By its Series G round at the start of 2026, it was worth $380 billion.

At the same time, each fundraising round diluted the Series B shares. The FTX estate told the bankruptcy court that Bankman-Fried's shares, which in 2022 represented 13.56% of Anthropic, represented 7.84% of the AI company in January 2024.

The Marshals Service sold Singh's and Ellison's shares to existing Anthropic investors sometime last year, according to the person with knowledge of the sale.

It's not clear when exactly the agency sold the shares, to which investors, how those investors were chosen, at what price the shares were sold, or how much money the government made in the sale. Depending on when they were sold in 2025, the combined shares could have been worth between $300 million and $1.1 billion, according to Sorenson, the UCLA professor. Rolfes, the PitchBook analyst, estimated a range between $250 million and $630 million, depending on the timing.

The Marshals Service declined to comment. A representative for the Justice Department said information about asset sales and victim compensation is confidential.

The revenue from the sale of Singh's and Ellison's Anthropic shares doesn't appear to have been transferred to the FTX estate as of the end of June this year, according to bankruptcy court filings from the estate, which continues to compensate victims and creditors.

The FTX estate received $638 million last year from assets seized by the Justice Department, according to the estate's annual report for 2025. It's not clear from the filings where those funds came from. The estate expects to receive about $400 million more from the government at some point in the future, according to the annual report. That would include proceeds from cryptocurrency and other investments made by Bankman-Fried.

One of Bankman-Fried's victims, Sunil Kavuri, told Business Insider that the Justice Department should use the proceeds of Ellison's and Singh's Anthropic shares to compensate victims. It would be "diabolical" for the government to hold onto the proceeds, he said.

Victims haven't actually been made whole, he said, because the bankruptcy court calculated the losses of FTX depositors using the time of FTX's bankruptcy declaration, when crypto prices were at a low ebb. Kavuri previously argued in bankruptcy court that the FTX estate should have held on to its Anthropic shares for longer to take advantage of the company's swift growth.

Representatives for the FTX estate declined to comment. The Justice Department spokesperson said the matter was ongoing, and that the DOJ "prioritizes victim compensation from forfeiture and takes all steps to ensure forfeited funds are provided to victims."

The government could end up just keeping the money, Bachner said.

"They've invested millions and millions and millions of dollars in prosecuting Bankman-Fried. And they want to get at least reimbursed for their costs, so sometimes they'll do that," he said. "It's really a unilateral government decision."

Jack Newsham and Katie Roof contributed reporting for this story.

Correction: August 31, 2026 — An earlier version of this story misstated the number of co-conspirators who testified against Sam Bankman-Fried. It was three.

Update: August 31, 2026 — This story has been updated to reflect Singh's role in the FTX fraud case.

Correction: October 27, 2026 — FTX bankruptcy court filings don't make clear where the Justice Department obtained the $638 million it transferred to the estate. A previous version of this story mischaracterized the filings.

• Anthropic's IPO paperwork shows the AI giant has some unusual approaches to governance.
• One of those is an agreement with a separate entity, Anthropic Founders LLC, controlled by seven of its cofounders.
• Those seven cofounders, including CEO Dario Amodei, can submit a secret list of successors to inherit their voting power.

Anthropic has devised an unusual way for its founders to pass on their power.

In its IPO filings, Anthropic outlined how seven of its cofounders, who collectively have 50.1% of voting control, will be able to create a ranked list of who they'd like to inherit their power in the event of their death or incapacitation.

What's particularly notable is that while the names would be shared with the other LLC members and Anthropic's Nominating Committee, they are otherwise contractually confidential and generally not public.

Anthropic CEO Dario Amodei, Anthropic President Daniela Amodei, and the five other LLC members will review and vote on whether to approve their fellow founders' proposed successors, but the names will likely never be seen by other shareholders. (The agreement allows for the publication only in narrow circumstances, such as a court order or other legal requirement.)

Under the agreement, which was first highlighted by Forbes, each member of the Founders LLC may name up to 10 potential successors, ranked in order of preference. The other LLC members must then vote to approve the secret lists.

While succession plans are not unusual, Anthropic's approach is unusual in how it ranks potential successors. It also underscores how, thanks to the Founders LLC, Anthropic's leaders will be able to maintain control as shareholders begin to offer input post-IPO.

Facebook's 2012 IPO paperwork suggested that if Mark Zuckerberg were in control of the company at the time of his death, then control could be transferred to someone he designated. (Meta has since tweaked this plan.) In 2020, Airbnb's three cofounders agreed to transfer their voting power to their other cofounders if one of them died or became disabled, but that arrangement (which still exists) only lasts for nine months.

The succession plan does not establish CEO succession. Anthropic's board of directors would handle such a chief executive appointment.

There are also some limitations to who the Anthropic founders can name on their lists.

Eight of their 10 potential successors are subject to eligibility requirements, including that the person named is a current or former employee of Anthropic or one of its subsidiaries, or has worked for "a 501(c)(3) tax-exempt organization or a national or international organization whose primary purpose is research, education, or the promotion of public benefit." In either case, the person must have worked there for at least three years.

If the person worked for Anthropic, they must have held a specific role, such as a member of the technical staff or the company's internal think tank, or another role approved by the board and a majority of the other members. If they worked outside Anthropic, they must have held a senior role at the nonprofit. These successors must also be US citizens, have permanent residency, or have valid work authorization.

The other two names, if a founder names all 10 potential successors, are not subject to the eligibility requirements, but they must also receive the required approval.

The succession plan doesn't last forever.

Under the agreement, a successor member generally would cease to be a member on the 10th anniversary of Anthropic's IPO. After that point, successor designees may be admitted only in limited circumstances during a transition period.

• Dario and Daniela Amodei said trust is key to operating Anthropic with seven cofounders.
• Operating with honesty and in good faith makes it easier to disagree, Dario Amodei said.
• Anthropic's S-1 filing outlined how the cofounders will steer the company after it goes public.

Many people told Dario and Daniela Amodei that starting Anthropic with seven cofounders was a terrible idea.

Yet, more than five years later, the duo said that setup is crucial to their success. The siblings made the comments in an episode of the podcast The Builders, released on Tuesday and recorded in July.

"They all have equal stakes in the company, and they all carry the vision. We have inherent trust in all of them," Dario Amodei, the CEO, said.

He said later in the episode that mutual trust is crucial to resolving his frequent disagreements with other cofounders on high-stakes questions, like what safeguards to apply to a model and how to lobby for regulations.

"These are all good-faith disagreements, but they always get resolved because there's always trust that everyone is trying to get to the right answer and everyone is being honest," he said, drawing a contrast to what he said was his previous experience working at OpenAI.

Daniela Amodei, Anthropic's president, said the focus on honesty also extends to how the leaders communicate with employees.

"Honesty and integrity and how we treat each other and our employees is also reflective of how we want the technology to be built," she said.

The podcast was released one day after Anthropic released its S-1 paperwork as it prepares to go public. As Henry Chandonnet wrote for Business Insider, the paperwork was chock full of language about honesty being central to the company's public persona.

Under a unique governance structure, as outlined in the S-1, the company's seven founders will control 50.1% of all shareholder votes, meaning the seven cofounders will have more voting power than all other shareholders combined. In addition to their voting power, the cofounders also hold significant stakes in the company and stand to earn billions. Each of the cofounders has pledged to donate 80% of their personal equity in the company to philanthropy.

Anthropic's highly anticipated S-1 paperwork also revealed key financial statistics: The company had a net loss of $42 billion and revenue of $4.6 billion last year, five customers drove around a quarter of revenue in the first nine months of the year, and the Amodeis together had total compensation packages valued at around $35 million last year. More than 90 pages of the S-1 outline risk factors, and Anthropic warned that a "catastrophic AI-related incident" could overwhelm the entire insurance industry.

The AI behemoth is expected to target a $1.5 to $2 trillion valuation when it debuts on the Nasdaq in November, potentially making it the largest IPO ever. Some investing experts have expressed concern that the market may struggle to absorb another blockbuster IPO.

• "The Big Short" investor Michael Burry says Anthropic's IPO filing has fueled his worries about AI.
• Burry said a dense web of AI deals has made the industry "catastrophically vulnerable."
• Writing on Substack, Burry described AI financing arrangements as "garish and audacious."

Michael Burry says Anthropic's IPO filing this week has reinforced his concerns that the entire AI ecosystem is at risk of disaster.

The investor of "The Big Short" fame wrote in a Substack post on Tuesday that the first in-depth look at Claude's parent company, ahead of what could be the biggest IPO in history, threw up a bunch of red flags.

Burry pointed to Anthropic's huge spending commitments, slowing revenue, and tangled ties.

He said that Anthropic has struck deals to buy about $537 billion of compute from the likes of Amazon, Alphabet, and Microsoft, and reached preliminary agreements worth hundreds of billions more with others such as SpaceXAI, putting its total commitments at about 12 times its current annual run-rate revenue.

"This is not stated clearly in the S-1, and in my view it is the most deserved headline from it," Burry wrote in a Tuesday Substack post.

He underscored how hyperscalers like Amazon and Alphabet act as Anthropic's "lenders, vendors, channel providers, and shareholders." He also outlined its deal with Broadcom, which involves Apollo and Blackstone and includes convertible notes and structured credit tranches.

"Just remember, the reason this is so complex is because too much debt is needed, and it must be unloaded from the balance sheets of those involved," Burry wrote.

"While some may see all the circularity and connectedness as security on the way up, and that is not wrong, it does leave the entire organism catastrophically vulnerable either on a pause or on the way down," he added.

He also described financing arrangements in the AI space as "garish and audacious," and said hyperscalers aren't generating enough cash to fund the AI buildout so they're resorting to "Byzantine" deals.

Burry said it was a "very similar" situation with vendor financing during the dot-com bubble, saying "people keep making the same mistakes with every new thing."

Burry's worries

In its filing, Anthropic said its revenue surpassed $16 billion for the first six months of this year, a 14-fold increase from the first half of 2025. Its operating loss narrowed by about two-thirds to below $700 million, and it generated nearly $4 billion in operating cash flow.

Burry initially commented on Anthropic's filing in a Substack chat thread on Monday.

He said that revenue was "slowing," citing Anthropic's claim that third-quarter revenue was 86 times the $197 million it generated in the fourth quarter of 2024, or about $16.9 billion.

That would mean Anthropic's revenue grew by about 46% between the second and third quarters of this year, after soaring 144% between the first and second quarters.

"Overall slowing of growth is not attributed to a decline in tokenmaxxing, but I did not expect that to be mentioned," Burry wrote.

Burry was referring to the trend of maximizing AI token consumption, which has fallen out of favor in recent months given its costs.

In another comment, Burry wrote that large language models (LLMs) like Claude are "improving" and "already being used more efficiently and that trend will continue, to the detriment of their growth rate."

Burry said he's focused on the complexity and fragility of the debt financing many AI deals, hurdles to installing new graphics chips shoring up demand for older chips, a potential crash in memory chip prices once the shortage eases, compression and small language models (SLMs) reducing the need for computing power, and the "fundamentals behind not just companies but the deals they are doing."

Responding to a subscriber who said that Big Tech companies are shifting from Anthropic's AI models to their own, and that tokens are rapidly being commoditized, Burry wrote: "Yes, that is what I hear."

Anthropic did not immediately respond to a request for comment on Burry's posts.

The contrarian investor shot to fame after his massive wager against the mid-2000s housing bubble was chronicled in the book and movie "The Big Short." He pivoted from running a hedge fund to writing about his personal investments on Substack last winter.

A recurring theme of his work since then has been deep skepticism of the AI boom. He's flagged slowing growth, overinvestment in microchips and data centers, aggressive accounting, hidden liabilities, and circular financing deals.

O Bank of America, a Blackstone, o Citi e o Morgan Stanley anunciaram nesta segunda-feira (26) uma operação de financiamento de US$ 63 bilhões para financiar a expansão da capacidade computacional de IA (inteligência artificial) da Anthropic, utilizando a tecnologia da Broadcom.

O financiamento tem como propósito a compra de aceleradores de IA e equipamentos de rede da Broadcom que serão alugadas à Anthropic. As entregas devem começar em 2027.

Os bancos Bank of America, Citi e Morgan Stanley serão responsáveis pela parcela do financiamento com garantia prioritária, enquanto a Blackstone será a única responsável pela parcela com garantia de segundo nível.

“A necessidade de computação para IA continua a crescer em um ritmo extraordinário. Isso valida a disponibilidade de capital para financiar esse crescimento e a posição única da Broadcom em viabilizá-lo”, afirmou o CEO da Broadcom, Hock Tan.

Mais cedo, a Anthropic registrou a proposta de oferta pública primária de ações (IPO, na sigla em inglês) na Comissão de Valores Mobiliários (SEC, na sigla em inglês) nos Estados Unidos.