More locations: San Francisco, CA, USA | New York, NY, USA
Staff must work from an office at least 25% of the time; some roles may require more.
Anthropic focuses on AI research to build reliable, interpretable, and steerable AI systems. Its main product, Claude, is an AI assistant designed to handle tasks at any scale for clients across industries, delivered through deployment and licensing along with specialized AI R&D services. Claude works by combining natural language processing, human feedback, reinforcement learning, and interpretability techniques to produce a capable, controllable AI assistant that can assist with a wide range of tasks. The company differentiates itself from competitors by prioritizing safety, transparency, and controllability—emphasizing reliability, interpretability of model behavior, and user-controlled steerability in its AI systems. Anthropic’s goal is to make AI systems that people can trust and efficiently use to improve operations and decision-making across sectors.
Company Size
5,001-10,000
Company Stage
Debt Financing
Total Funding
$182.8B
Headquarters
San Francisco, California
Founded
2021
See people who can refer or advise you
Help us improve and share your feedback! Did you find this helpful?
Flexible Work Hours
Paid Vacation
Parental Leave
Hybrid Work Options
Company Equity
Anthropic has launched Sonnet 5.5, its latest mid-tier AI model, positioning it as a faster and cheaper alternative to its predecessor. The company claims Sonnet 5.5 operates 30% faster than Sonnet 5 with significantly reduced token consumption, making it ideal for everyday tasks like coding and document creation. Sonnet 5.5 sits below Opus in Anthropic's model hierarchy but demonstrates superior performance in agentic coding tasks. The model can spawn multiple agents without exceeding cost limits, according to company benchmarks. Notably, Sonnet 5.5 features cyber capabilities comparable to Opus 5, making it the first Sonnet model subject to the same cyber safeguards as Fable and Opus. Anthropic also announced plans to release an updated version of Haiku, its smallest model, in the coming weeks. The release follows recent model launches from OpenAI and Meta.
Microsoft has redesigned Copilot to compete directly with Anthropic's Claude, consolidating years of separate AI products into a single app. The update was announced on 25 September. The redesigned Copilot features three sections: Home, which combines chat with task delegation and embeds Word, Excel and PowerPoint; Code, enabling users to build apps using natural language; and Autopilot, allowing creation of cloud-based agents for recurring tasks. The move comes as Microsoft's stock has trailed megacap peers this year. Company executives acknowledged they are playing catch-up with OpenAI and Anthropic, which have bundled chat, coding and agent capabilities. Microsoft is differentiating on enterprise trust rather than raw capability, citing security concerns with competitors' approaches.
Amazon shares dipped to $248.515 on 25 September after Anthropic announced a seven-year, $11.6 billion cloud infrastructure commitment to Akamai. The deal covers CPU workloads and includes a warrant potentially giving Anthropic a 5% stake in Akamai, with an additional $9 billion in possible future business. The arrangement does not indicate Anthropic is moving existing workloads away from Amazon Web Services. Amazon reported $53.4 billion in non-operating pre-tax other income in the second quarter, primarily from its Anthropic investments. Whilst Amazon benefits from Anthropic's rising valuation through its investment stake, the cloud infrastructure contract highlights that backing an AI company does not guarantee capturing its entire computing budget.
AskNicely launches Claude Connector for live CX feedback data. AskNicely has integrated its CX platform with Anthropic's Claude, letting businesses query live NPS, CSAT and review data conversationally alongside other enterprise data sources. What happened. AskNicely has launched a new Claude Connector, integrating its customer experience platform directly with Anthropic's Claude AI assistant. The connector allows businesses to query live Net Promoter Score (NPS), customer satisfaction (CSAT) and review data conversationally within Claude, alongside other operational and business data sources they already connect to the assistant. According to reporting from CustomerThink and Business Insider, the integration is designed to replace manual, spreadsheet-driven feedback analysis with natural-language queries that return answers in minutes rather than days. Fitness brand solidcore is cited as an early adopter, using the connector to interrogate customer feedback trends without waiting on dedicated analyst time or static dashboards. The launch positions AskNicely's feedback data as one of several enterprise data sources Claude can now reason over in a single conversation, rather than requiring users to move between separate CX reporting tools and AI assistants. Why it matters. This is fundamentally a story about how AI assistants are becoming the interface layer for enterprise data, rather than a new destination in their own right. By making CX metrics queryable inside Claude, AskNicely is betting that feedback data will be more useful sitting alongside sales, operations or finance data in a general-purpose assistant than living in a standalone CX dashboard that few non-specialists ever open. For experience and transformation leaders, the shift matters because it lowers the barrier between insight and action. Feedback trends that once required a CX analyst to compile a report can now, in principle, be surfaced by any manager who knows how to ask the right question - changing who gets to use customer data, and how quickly. The renascence take. The real story here isn't the technical plumbing of a connector - it's what happens to an organisation's relationship with feedback once the friction of asking for it disappears. Most CX programmes don't fail because the data is bad; they fail because insight arrives too late to change a decision, and too narrowly to reach the people who could act on it. Putting NPS and CSAT inside a conversational assistant doesn't automatically fix that - it just removes the excuse. The operators who benefit will be the ones who pair this kind of access with clear ownership of what "good" looks like, so a two-minute query produces a decision, not just a faster version of the same unread report. Without that discipline, instant access to feedback data risks becoming instant access to noise. This briefing was written by its Newsdesk, synthesising reporting from the outlets below. Follow the links for the original coverage. Questions Renascence get on this topic. It integrates AskNicely's customer experience platform with Anthropic's Claude AI assistant, letting businesses ask natural-language questions about live NPS, CSAT and review data instead of relying on spreadsheets or static dashboards. More in Customer Experience Stay ahead of CX Get the signal, not the noise. The stories shaping customer experience - plus the Journal and Experience Loom - in your inbox.
Basecamp Research raises $140m to train AI on the world's rarest DNA. Basecamp Research, a London-based start-up that trains artificial intelligence on DNA rather than language, has raised $140m from investors including Anthropic, Nvidia and the UK's sovereign AI fund. The company said the money will pay for data collectors to travel from Chile to the high Arctic in search of the planet's rarest genetic code, which will be used to train a new generation of its Eden AI models. Basecamp did not disclose its valuation, but said it was "very close" to $1bn. Free newsletters The stories that matter to UK business, straight to your inbox. The Eden models run on top of the "Trillion Gene Atlas", which Basecamp claims is the world's largest proprietary biological AI training dataset. Glen Gowers, co-founder and chief executive, said the shortage of biological data to train AI was an industry-wide problem that could limit the speed of advances in AI-designed treatments. "The problem with biology is you don't have internet-scale data available," he said. "What we're bringing to the table as an AI company is the missing data layer." Others in the sector have pointed to the same gap: AI drug-design start-up deepmirror recently secured a £1m Sovereign AI grant to expand its molecular dataset for the same reason. Basecamp was founded in 2019 by Gowers and Oli Vince, who met at the University of Oxford. On an expedition to Iceland, the pair became the first to sequence DNA completely off-grid. The company has since conducted about 150 expeditions around the world. "It's shovels and soil," Gowers said. "We go to the far-flung corners of the planet, the tops of ice caps, the bottom of the ocean, volcanoes and islands." Alongside deepening existing partnerships in countries including Cameroon, Malawi and Chile, the company said the investment will allow it to conduct research in Oceania and the high Arctic. By training AI on the data it collects, Basecamp aims to develop new medicines and design treatments faster than traditional methods allow. In the immediate future it will focus on cell therapies, an approach in which a patient's cells are effectively reprogrammed. "We believe the future of medicine lies in reprogramming the body to repair itself," Gowers said. The company said its models allow it to design long and complex DNA sequences faster, cheaper and more accurately than existing cell therapies. The funding will also be used to advance a pipeline of AI-designed therapeutics. Two therapies from its Boston laboratory are already showing promising results in mice, according to the company. Basecamp has offices in Clerkenwell, London, and in Boston. The state-backed sovereign AI fund's participation makes Basecamp the sixth company it has invested in directly. The £500m Sovereign AI fund typically makes equity investments of between £1m and £10m. The government launched the Sovereign AI programme in April, with drug discovery among the sectors covered by its first round of backing. Venture capitalist Suzanne Ashman was later appointed managing partner of the £500m fund. Lord Vallance of Balham, the chairman of the government's AI taskforce, said: "AI will be a powerful tool for turning scientific discovery into better care for patients, faster than ever before. "By accelerating the design of advanced therapies and making them simpler to deliver, innovations like these could bring transformative treatments to many more patients in need."