AI research startup According to company documents obtained by TechCrunch, Anthropic aims to raise up to $5 billion over the next two years to compete with rival OpenAI and enter more than a dozen major industries.
Anthropic’s Series C funding round pitch deck reveals these and other long-term goals for the company founded in 2020 by former OpenAI researchers.
Anthropic said in a document that it plans to build a “frontier model” (provisionally dubbed “Claude-Next”) that is ten times more powerful than the most powerful AI today. expects to spend $1 billion over the next 18 months. .
The Information reported in early March that Anthropic was seeking to raise $300 million at a $4.1 billion valuation, bringing the total raised to $1.3 billion. Deck confirms that target number, but only half was raised from “confidential investors” at the time of writing the document.
Anthropic describes Frontier Models as “next-generation algorithms for AI self-learning,” referring to an AI training technique it developed called “constructive AI.” Broadly speaking, constructive AI seeks to provide a way to adapt AI to human intent. In other words, let the system answer your questions and perform your tasks according to a set of simple guidelines.
Anthropic estimates that its frontier model will require 10^25 FLOPs, or on the order of floating point operations. This is several orders of magnitude larger than the current largest model. Of course, hWhy this translates into computation time depends on the speed and scale of the system performing the computation. The company implies that it relies on clusters with “tens of thousands of GPUs.”
You can use this frontier model to build virtual assistants that can answer emails, conduct research, generate art and books, and more. Some of these have already been tasted with GPT-4 and other large language models.
“These models could start automating large parts of the economy,” reads Pitchdeck. “We believe the companies training the best of his 2025/26 models are too far ahead to catch up in subsequent cycles.”
Frontier Model is the successor to Anthropic’s chatbot, Claude, and can be commanded to perform a variety of tasks, including searching through documents, summarizing, writing and coding, and answering questions about specific topics. In these respects, it is similar to his ChatGPT on OpenAI. However, Anthropic claims Claude is “much less likely to produce harmful output”, “easier to talk to” and “more maneuverable” thanks to its constitutional AI.
Anthropic released Claude commercially in March following a closed beta last year, allowing organizations to request access. Beta users and potential customers include the following industries (asterisks indicate human involvement to monitor the model):
- Summary and Analysis of Legal Documents*
- Medical patient recording and analysis*
- Customer service email and chat
- Coding model for consumers and B2B
- Productivity-related search, document editing, content generation*
- Chatbot for public Q&A and advice
- Search using natural language responses
- HR tasks such as job descriptions and interview analysis*
- therapy and coaching
- Virtual Assistant*
- All Levels of Education*
Former VP of Research at OpenAI, Dario Amodei, launched Anthropic as a public benefit corporation in 2021, bringing along many OpenAI employees, including former OpenAI policy leader Jack Clark. Amodei split from OpenAI after disagreements over the company’s direction, the startup’s increasingly commercial focus.
Anthropic currently competes with OpenAI and startups like Cohere and AI21 Labs. All of these startups are developing and commercializing their own text-generating (and sometimes image-generating) AI systems.
The pitch deck notes that Alameda Research Ventures, a sister company of Sam Bankman-Fried’s failed cryptocurrency startup FTX, is a “silent investor” with a “non-voting” stake in Anthropic and a $580 million stake in Anthropic. It announced that it spearheaded a $100 million Series B round. Anthropic expects Alameda shares to be disposed of in bankruptcy proceedings within the next few years.
Google is also an investor in Anthropic, pledging $300 million to acquire a 10% stake in Anthropic. 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. ”
Other backers of Anthropic include James McClave, Facebook and Asana co-founder Dustin Moskovitz, former Google CEO Eric Schmidt, and founder Skype engineer Jaan Tallinn.
