On Monday, Amazon announced they will invest up to $4bn (£3.3 billion) in Anthropic and take a minority stake in the AI firm.
Amazon said Anthropic’s AI platform called Claude would boost Alexa’s conversational powers and improve customer experience. This “strategic collaboration” stated by amazon, will bring together industry-leading technology and expertise; Amazon engineers will build with Anthropic models to deploy state-of-the-art AI systems.
With Wednesday’s announcement came early access to the new Amazon Bedrock, a platform to develop and customise generative AI applications using foundation models.
Amazon chief executive, Andy Jassy said: “We have tremendous respect for Anthropic’s team and foundation models, and believe we can help improve many customer experiences, short- and long-term, through our deeper collaboration”.
Anthropic was founded in 2021 by the former members of OpenAI, Dario Amodei and Daniela Amodei.
Amazon provides computing services and rents its massive stores of power at data centres to firms. The collaborative investment will allow Anthropic to draw on this immense computing power, and Amazon will be able to use Anthropic's latest AI model, Claude 2. Anthropic will use Amazon's Trainium and Inferentia chips - designed specifically for generative AI applications - to develop its new AI models.
Claude 2 performs better than Claude 1.3, with a safer output of information, better global availability, and overall coding performance. It scored 76.5 percent, up from 73.0 percent on the Bar exam, scored 71.2 percent up from 56.0 percent on the Codex HumanEval coding test, and scored 88.0% up from 85.2% on the GSM8k math test.
AI Race
Major tech firms are currently racing to invest in AI to improve their platforms. 3 years ago, Microsoft’s invested $10 billion in OpenAI, receiving 75 percent of OpenAI's profits.
Several more cloud providers are expected to announce their own investments in generative AI. Google invested $300m in Anthropic last year, and the company said it would train its models on Google’s chips and use its cloud services.
Anthropic disclosed Claude 2 in July, which rivalls ChatGPT in the beginning of an AI race. Ethan Simmons, the managing partner of cloud consultancy and Pinnacle Technology Partners said, “What we are seeing is the early stages of an AI race as to who is going to be able to come up with a monetizable AI product. [...] The Amazon investment in Anthropic will help create monetizable AI products faster for AWS and its partners".
Anthropic recently spoke on the ethical responsibility of AI. They said they support a set of voluntary safety commitments led by the White House.
In May, a meeting on the ethics of AI was held which brought together the leaders of Anthropic, DeepMind, and OpenAI with the UK Prime Minister, Rishi Sunak. Potential risks from AI were discussed, including disinformation, national security, "existential threats", and regulatory action.
Edited by Anwen Venn
Share This Post On
