Itās been a quite, end of summer week in the world of generative AI but we had a few interesting announcements, including Anthropic launching Claude Enterprise, rumours that the next generation of Amazonās Alexa devices will be powered by Claude and Ilya Sutskeverās new AI company raising $1bn of funding.
In ethics news, the US, EU, UK and other markets announced that they have all signed a legally enforceable AI treaty and there are some good long reads from The Washington Post on OpenAIās SearchGPT and The New York Times about a musician who gamed the streaming system with generated music to win royalty payments of $10m š¤Æ.
Anthropic launches Claude Enterprise Plan with GitHub integration and 500K context window
Itās great to see Anthropic launch Claude Enterprise as it gives businesses access to Claude 3.5 Sonnet, arguably the best model available right now. Iām also a big fan of the project organisation in Claude and also their Artefacts feature. For me, itās the best user experience of all the frontier GenAI models right now and the reason I switched from using ChatGPT earlier this year.
The GitHub integration is also a really nice touch, allowing Claude Enterprise to compete with GitHubās Copilot.
Claude Enterprise comes with the typical Enterprise features weāre now seeing from offerings on the market such as no training on Enterprise data, single sign on and audit logs. Iām looking forward to seeing what businesses can do with it!
Amazonās new Alexa voice assistant will use Claude AI
Amazon took a big part in Anthropicās funding round in September last year, investing $4bn in the company. At the time it looked like Amazon was just looking to get a foothold on the frontier of generative AI, but now it looks like that investment was more strategic - Anthropicās Claude model will power the next generation of Alexa voice assistants.
According to reports, Amazon have been struggling since last year to get their own generative AI models to reliably power what theyāre calling āRemarkable Alexaā and have now decided to power their next assistant with Anthropicās Claude model.
More will be revealed at Amazonās annual devices and services event later this month and there could be a subscription cost attached to the new version of Alexa.
OpenAI co-founder Sutskever's new safety-focused AI startup SSI raises $1 billion
It was back in June that Ilya Sutskever announced that he was launching his new AI startup SSI (Safe Superintelligence) and in just 3 months theyāve raised $1bn at a valuation of $5bn. The company currently only has 10 employees including co-founders Daniel Gross and Daniel Levy and will be using the majority of the funds raised to acquire computing power to train their AI models.
Not much more information has been shared, but the list of investors is impressive, including Andreessen Horowitz and Sequoia Capital.
Tesla plans robotaxi reveal at California Warner Bros. Studios
Tesla will be unveiling their long-awaited Robotaxi on October 10th at Warner Bros. Studios in California. The studio lot if likely a good venue as there are dozens of sound stages with fake suburban towns, allowing Tesla to train Robotaxi on the layout and conduct demos for journalists.
It will be an interesting event as Elon Musk has been hyping Robotaxi for years, but Tesla has so far failed to deliver on their repeated promises of Full Self Driving vehicles, which is the core technology Robotaxi will be built around. Iām looking forward to seeing the driverless design of the vehicles at the very least!
AI Ethics News
New York Times writer exposes how AI models can be fooled by invisible text on websites
Sam Altman indicated it's impossible to create ChatGPT without copyrighted material
Long Reads
The Washington Post - Few have tried OpenAIās Google killer. Hereās what they think.
The New York Times - The Bands and the Fans Were Fake. The $10 Million Was Real.
āThe future is already here, itās just not evenly distributed.ā
William Gibson