A week in Generative AI: OpenAI, Character.ai and Jony Ive
A week in Generative AI: OpenAI, Character.ai and Jony Ive
News for the week ending 1st October 2023
And I thought last week was a busy week for generative AI news… this week sees numerous big announcements from OpenAI, plus new from Meta, Character.ai, Amazon & Anthropic and Microsoft. Oh, and Jony Ive could be back to help Sam Altman some AI centric hardware…. nothing to see here 🙈.
OpenAI announces some big updates (and some rumours…)
This week OpenAI announced DALL-E 3 and its integration into ChatGPT. It also announced the return of Web Browsing. So now (or very soon!) ChatGPT will be able to see, hear, speak, output images and have up-to-date knowledge. That’s a pretty big week all round!
There are also rumours that what we’re seeing publicly is just a flavour of what’s to come. Their ‘Arrakis’ model sounds like a GPT 4.5 with a large number of improvements and is an ‘everything-to-everything’ model.
Then there’s AGI rumours and this tweet from Sam Altman. Interesting times ahead!
ChatGPT can now see, hear, and speak: Source
Web Browsing: Source
Rumours: Source
Generative AI as the foundation of future operating systems
I couldn’t have put this tweet from Andrej Kaparthy better and is totally in line with what we’ll see with generative AI becoming the foundation of the next computing paradigm.
Meta announces a sweet of new AI features
So much to unpack here from Meta, but the big thing for me is the idea of generative AI models based on and endorsed by celebrities. God knows how much Meta is paying them but giving people the opportunity to have a conversation with their heroes could be a big money spinner. Very much like an official Character.ai model. Speaking of which…
Character.ai in Early Talks for Funding at More Than $5 Billion Valuation
If this funding leads to Character.AI being able to scale faster then they already are then I think we’ll see if quickly become the next TikTok and a truly scaled generative AI entertainment platform.
Jony Ive and OpenAI CEO Raise $1B to Design the 'iPhone of AI'
This could be an amazing collaboration but I wouldn’t expect anything to come from it for at least 2-3 years, if at all. Meanwhile, we have Humane’s ‘AI Pin’ debuting on the Paris runway. Hardware will see a shakeup but I just don’t think we’ll be able to get away from needing screens until they’re strapped to our heads.
Amazon to invest as much as $4 billion in AI startup Anthropic
That’s a lot of money! I was slightly surprised to see this from Amazon at first, but then I realised that they’re the only one of the current big tech firms that didn’t have an AI play of their own. So, makes a lot of sense.
Microsoft 365 Copilot Release Date Set for November
I don’t think this will be as impressive as it makes out to be at first, but is a really important first step towards the future computing paradigm mentioned above that I expect we’ll see.
OpenAI is reportedly raising funds at a valuation of $80 billion to $90 billion
So much cash monies flying around this space now! I previous did some rough maths and worked out that it would take c.100 years for OpenAI to pay back Microsoft’s $13bn investment, even with revenues of $1bn a year. But, this could just be a nice way for the company to let employees start benefiting from the shares they own and the current hype-cycle.
This week’s AI Ethics news
The writers strike is over; here's how AI negotiations shook out
Oops! Google Search caught publicly indexing users’ conversations with Bard AI
Deepfake election risks trigger EU call for more generative AI safeguards
Google adds a switch for publishers to opt out of becoming AI training data
This week’s long reads
Scientific America - Artificial Intelligence Could Finally Let Us Talk with Animals
“The future is already here, it’s just not evenly distributed.“
William Gibson