Anthropic, a high-profile AI startup co-founded by former OpenAI employees, has begun giving partners access to AI text generation models.
The first commercial venture to announce the integration of the Anthropic model is Robin AI, a legal tech startup that has raised over $13 million from investors including Plural, Episode 1 and Google Black Founders Fund. Poe, his experimental chatbot app for iOS and Android on Quora, uses the Anthropic model and is currently unmonetized.
Robin CEO Richard Robinson provided few details about the Anthropic relationship, but told TechCrunch that Robin fine-tuned the Anthropic model based on a dataset of legal documents and drafted and negotiated the contract. Told.
“We are very fortunate to be Anthropic’s launch partner in the legal sector. match. The state-of-the-art model is ‘hallucination,'” Robinson said in a statement.
Robin AI is one of the first commercial ventures using the Anthropic model.
Anthropic has been relatively quiet about its plans to commercialize it in the generative text AI space, preferring instead to focus on academic research. But late last year, the company launched a closed beta version of an AI system called Claude, which is similar to OpenAI’s ChatGPT and appears to improve on the original in key ways. It was created using a technique called “constitutional AI” that we developed. This technique aims to provide a “principles-based” approach to aligning AI systems with human intent. Use a ChatGPT-like AI to answer questions using a simple set of principles (e.g. give harmful advice) as a guideline.
Claude’s preliminary impressions were good. However, like ChatGPT, the system has limitations such as providing dangerous answers to questions (such as how to make stimulants at home) or making inconsistent, factually incorrect statements. had.
Not sure if model Robin uses teeth Claude or its derivatives — neither Robin nor Anthropic. Even after repeated offers, Anthropic reveals how many partners it currently works with (or how it came to work with them) and how many models it plans to open up for commercial use. I didn’t do it.
But make no mistake, Anthropic is under pressure from investors to recoup hundreds of millions of dollars spent on AI technology.
Most recently, Google invested $300 million in Anthropic, pledging to acquire a 10% stake in the company. Under the terms of the deal, first reported by the Financial Times, Anthropic agreed to make Google Cloud its “preferred cloud provider,” and the two companies “co-developed.”[ing] AI computing system. ”
Curiously, Anthropic was not founded with a profit-seeking mission. Dario Amodei, former vice president of research at OpenAI, launched the company as a public benefit corporation in 2021, bringing with him many of his OpenAI employees, including former OpenAI public policy leader Jack Clark. Amodei split from OpenAI after disagreements over the company’s direction, the startup’s increasingly commercial focus.
However, AI systems are expensive to develop and maintain. Inflated costs prompted Anthropic to seek outside help. This includes his $580 million from a group of investors including his infamous FTX founder Sam Bankman-Fried, Caroline Ellison, Jim McClave, Nishad Singh, Jaan Tallinn, and the Center for Emerging Risk Research. Includes a $10,000 tranche.
It’s not yet clear whether partnerships with startups, and investments in big tech, mean a shift in Anthropic’s priorities.but what teeth What is clear is that the company believes its technology is differentiated to compete with rivals such as OpenAI, Cohere and AI21 Labs. All of these companies offer paid access to their text generation AI via APIs.
