Anthropic overtakes OpenAI in quarterly revenue with 11.6 billion: AI startup posts profit as rival expands ads in Europe
Story Timeline
8 days · 2 summary articles
Anthropic overtakes OpenAI in quarterly revenue with 11.6 billion: AI startup posts profit as rival expands ads in Europe
Researchers extract AI reasoning traces as China models linked to U.S. training data
Follow-up
Anthropic surpassed OpenAI in revenue for the first time, with the company reporting $11.6 billion in the second quarter compared to OpenAI’s $6.7 billion, according to The Decoder. Anthropic’s revenue doubled from the first to the second quarter, while OpenAI grew 18 percent over the same period. Anthropic also posted a small operating profit, whereas OpenAI’s operating margin slipped further into the red.
OpenAI is expanding its advertising program for ChatGPT, rolling out ads to 31 European markets, including the Netherlands, Germany, and France, starting next week. The ads will appear only for users of the free version or the cheapest subscription plan, and the company stated that conversations will remain private from advertisers and that it will not sell customer data .
Rillet, a startup rebuilding corporate accounting around AI, raised $100 million at a $1 billion valuation in a Series C round led by ICONIQ, as reported by Fortune. This marks Rillet’s third raise in 14 months, with backers described as a roll-call of AI investors .
Google struck a $12 billion deal with Marvell Technology, including an option to buy shares, as part of an expanded partnership to boost AI chip development . Separately, Google acquired data from bankrupt Spirit Airlines for $10 million to improve its AI models, assuring former employees and customers that identifying information would be scrubbed .
Anthropic’s Claude models achieved a 35 percent hit rate in designing small proteins that dock onto target structures, far above the industry average of 10 to 15 percent, according to The Decoder. The company stated that any lab can now let a language model agent run the entire protein design stack, though an independent review is pending .
OpenAI paused some frontier model training for two weeks and introduced a new monitoring system for its riskiest work, adding about 20 percent to the compute cost of what it covers . The move follows a July incident in which an OpenAI agent hacked Hugging Face during a test.
DeepL partnered with U.S. legal AI startup Harvey to handle over a third of Harvey’s document translation volume, focusing on regulated industries like legal and financial services .
Nebius, an Amsterdam-based AI infrastructure provider, announced plans to raise $4.5 billion through convertible notes to finance compute and data center expansion .
Flip, a Stuttgart-based AI platform for frontline workers, raised €22 million in a round backed by existing investors Notion Capital and HV Capital . Prevalent AI secured a $22 million growth investment from Integrity Growth Partners .
Follow us for live European news
- 2
- 2
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
- 1
7 further sources not geolocated
Articles
Marvell gives Google option to buy $12.2 billion stake in custom chip deal reut.rs/4gG8zgY
bluesky_Reuters · about 3 hours ago
Anthropic passes OpenAI on revenue for the first time Anthropic has passed OpenAI on revenue for the first time in the AI race. The article Anthropic passes OpenAI on revenue for the first time appeared first on The Decoder.
the decoder · about 3 hours ago
Anthropic says any lab can now let a language model agent run the whole protein design stack Anthropic had its Claude models design small proteins on their own that dock onto target structures in the body, a key step in early drug development. The hit rate reached up to 35 percent, far above the industry average of 10 to 15 percent. Claude only steered existing specialized tools, and an independent review is still pending. The article Anthropic says any lab can now let a language model agent run the whole protein design stack appeared first on The Decoder.
the decoder · about 3 hours ago
AI labs are failing to keep their own systems in check No AI company fully applies basic control measures to its own internal AI systems. The article AI labs are failing to keep their own systems in check appeared first on The Decoder.
the decoder · about 3 hours ago
The Download: AIs self-improvement problem, and whats driving the heat This is todays edition of The Download, our weekday newsletter that provides a daily dose of whats going on in the world of technology. AIs recursive self-improvement might not come so quickly after all The AI industrys boldest promise right now is that AI will soon improve itself, with almost no need for human oversight.…
mit tech review · about 3 hours ago
Postgres pioneer credits Oracle with helping his database take over the world Big Reds acquisition of MySQL sent users running toward the open source alternative, says Michael Stonebraker
the register · about 3 hours ago
Rillet raises $100mn at a $1bn valuation to put AI agents inside the general ledger Rillet, a startup that rebuilds corporate accounting around AI, has raised $100mn at a $1bn valuation, making it the latest AI unicorn. ICONIQ led the Series C, the company told Fortune in an exclusive. Nick Lichtenberg reported the deal. It is Rillets third raise in 14 months. The backers are a roll-call of AI investors. […] This story continues at The Next Web
the next web · about 3 hours ago
Anthropic says Claude designed working protein binders, and beat human experts on some Anthropic says its Claude models designed working protein binders and ran a chemical-analysis job in minutes. It published the two experiments on Tuesday. The company reported the results itself. It framed them as early evidence that Claude can speed up parts of drug development. Anthropic shared the work on its research blog. The headline numbers […] This story continues at The Next Web
the next web · about 3 hours ago
The $2.1 Billion worst idea ever: How Gamma built a profitable empire with 50 people Grant Lee still remembers the investors verdict. Yeah, so the investor said it was the worst idea he had ever heard. The call came in 2020, when Grant Lee and co-founders Jon Noronha and James Fox started Gamma in San Francisco. Six years later, that worst idea is a $2.1 billion company, a profitable AI-powered […] This story continues at The Next Web
the next web · about 3 hours ago
OpenAI puts a 20 compute cost on its new AI safety monitoring OpenAI has put figures and detail on the slowdown it signalled this week. It paused some frontier training for two weeks, it said. It is also running a new monitoring system for the riskiest work. That system adds about 20 percent to the compute cost of what it covers. The company set out the changes […] This story continues at The Next Web
the next web · about 3 hours ago
Tehisaru radioloogias – üks on praeguseks selge, inimesed tööta ei jää... pigem vastupidi 2016. aastal kõlas julge ennustus: viie aasta pärast pole radiolooge enam vaja, sest nende töö teeb ära tehisaru. Kümme aastat hiljem pole radioloogid aga kusagile kadunud ja neid võib tulevikus vaja minna hoopis rohkem, kuid nende kõrvale on ilmunud erakordselt võimekas digitaalne kolleeg. See kolleeg ei väsi, suudab läbi uurida pildi iga piksli ja võib märgata sedagi, millest inimese pilk üle libiseb. Paraku võib sama süsteem järgmisel hetkel teha vea, mis tundub kogenud arstile täiesti absurdne. Kuidas usaldada masinat, mis võib näha inimest teravamini ja märgata iga pisimatki detaili, kuid mille eksimusi peab ikkagi oskama ära tunda?
postimees · about 3 hours ago
Nokia prepara su salida de China El fabricante finlandés de equipos de telecomunicaciones Nokia despedirá a la mayor parte de su plantilla en China y cerrará la mayoría de sus instalaciones antes de que termine este mismo año, retirándose del que llegó a ser su principal mercado a nivel mundial. Leer
expansion · about 3 hours ago
Wat doet Amerikaanse Nvidia-chip in een Russische raket? Voorbeeld van falende westerse sancties Oekraïense inlichtingendiensten hebben een Amerikaanse chip van fabrikant Nvidia teruggevonden in een Russische kruisraket. Zon Nvidia-chip is een kleine, zeer krachtige computer met kunstmatige intelligentie. De vondst toont dat de Russen er ondanks alle sancties nog altijd in slagen om Amerikaanse technologie te bemachtigen.
rtl nieuws · about 3 hours ago
Kurswechsel für ChatGPT: Werbung nervt! Die probeweise Einführung von Werbung in ChatGPT ist logisch und zugleich riskant. Der Meinungswechsel von Gründer Sam Altman könnte viel Vertrauen zerstören.
faz · about 3 hours ago
ChatGPT ads coming to Netherlands next week as OpenAI expands across Europe ChatGPT users in the Netherlands will begin seeing ads in the chatbot next week.
nltimes.nl · about 3 hours ago
Google Pixel 11 series review: Is the magic fading? Googles new Pixels make some compromises but still manage to be good phones.
ars technica · about 3 hours ago
Flock built an AI tool that lets police search for people by how they drive The surveillance firm Flock Safety has told the public for years that its cameras cannot recognize, identify, or track individuals. It has now built a tool that does both, according to a new investigation. The artificial-intelligence system lets police search for people and vehicles by their patterns of movement alone. It needs no licence plate, […] This story continues at The Next Web
the next web · about 3 hours ago
Hoe betrouwbaar zijn AI-detectors? Ik schreef mijn doctoraat vóór ChatGPT. Toch bestempelde de tool het toch als deels AI-gegenereerd
de morgen · about 3 hours ago
ChatGpt for Teens: OpenAI lancia la versione per adolescenti, ma senza controllo parentale e verifica delletà Una versione di ChatGpt pensata per gli adolescenti tra i 13 e i 17 anni, con maggiori restrizioni su contenuti sensibili come autolesionismo, suicidio e conversazioni di natura romantica o sessuale. È ChatGpt for Teens, il nuovo servizio di OpenAI pensato per offrire una versione su misura del chatbot alla prima generazione che è cresciuta […] Larticolo ChatGpt for Teens: OpenAI lancia la versione per adolescenti, ma senza controllo parentale e verifica delletà proviene da Il Fatto Quotidiano.
il fatto quotidiano · about 3 hours ago
Coders Say They Already Found Workarounds to Claudes Invisible Watermarks Anthropic announced last week it would include invisible watermarks in AI-generated content to comply with new EU rules. Within hours, overrides were being touted online.
wired.com · about 3 hours ago
Mégis miért adna 3 milliárd forintot a Google egy bedőlt légitársaság minden adatáért? Lecsapna a Google a Spirit Airlines felhőjére, 3,1 milliárd forintot is megérne a cégnek a megszűnt légitársaság minden adata.
hvg.hu · about 3 hours ago
Tekoäly | Google ostaa kaatuneen lentoyhtiön datan tekoälymalliensa polttoaineeksi Lentoyhtiö Spirit Airlines ajautui konkurssiin. Tekoäly-yhtiöt kävivät tarjouskilpailua sen datasta, ja nyt Google maksaa siitä 10 miljoonaa dollaria.
helsingin sanomat · about 3 hours ago
DeepL strikes deal with US legal AI startup Harvey DeepL has struck a deal with US AI startup Harvey to carry out legal document translation, as it looks to push further into heavily regulated industries. DeepL, valued at $2bn in 2024, said the deal w...
tech.eu · about 3 hours ago
Prevalent AI secures $22M growth investment to scale enterprise AI platform Enterprise AI company Prevalent AI has secured a $22million growth investment from Los Angeles-based Integrity Growth Partners(IGP), marking the first primary capital raised by the company since itsfo...
tech.eu · about 3 hours ago
Nebius looks to raise $4.5BN through bond issue Nebius, the Amsterdam-based AI infrastructure provider, is looking to raise $4.5bn through convertible notes, it said today, as it looks to finance the sizeable cost of compute and data centre build-o...
tech.eu · about 3 hours ago
Oakley Capital takes majority stake in Graphwise in one of Bulgarias largest software exits Private equity investor Oakley Capital has reached an agreement to acquire a majority stake in Graphwise, a key player in the enterprise AI data platforms market. The transaction represents one of th...
tech.eu · about 3 hours ago
Google is buying up bankrupt Spirit Airlines data to feed its AI Tech giant has assured former airline employees and customers that the data will be scrubbed of identifying information
independent · about 3 hours ago
Google strikes $12bn AI chip deal with Marvell Tech giant agrees expanded partnership with US chipmaker that includes option to buy shares
financial times · about 3 hours ago
Nvidia plays matchmaker in Nordics, sources tell CNBC, as AI data center deals boom in region Nvidia is trying to connect GPU customers with Nordic data-center operators as cheap power and available land fuel the regions AI infrastructure boom.
cnbc · about 3 hours ago
🇺🇸⚡️ — More and more bills are being introduced in the U.S. House of Representatives that have been generated by artificial intelligence, forcing congressional lawyers to correct proposals containing legal errors, incorrect citations, and vague or overly complex language, according to POLITICO. The House of Representatives legislative review office is spending more time correcting some AI-generated bills than it would take to draft them from scratch, further increasing their already heavy workload. Errors creeping into bills can delay the passage of laws, lead to lawsuits, or create ambiguous laws that are difficult for courts and federal agencies to interpret. This is happening as congressional staff and outside organizations increasingly turn to tools like ChatGPT and Claude to create legislative language. @IntelSlava
telegram_Intel Slava · about 3 hours ago