This is what I found interesting around AI this week, ending August 30, 2024.
TLDR:
Midjourney is Getting Into Hardware
Custom Gems and Improved Image Generation with Imagen 3 in Gemini
The controversial California AI bill that has divided the tech world
OpenAI and Anthropic will share their models with the US government
100M Token Context Windows
Midjourney is Getting Into Hardware
Not sure what that means to be honest with you, but they tweeted yesterday that they are getting into hardware and if you are interested to join their team in San Francisco, to email them. Are they going to create chips? Are they creating full GPUs? Are they creating their own cloud? It will be interested in seeing what comes out of this. Check out their tweet here.
Custom Gems and improved image generation with Imagen 3 in Gemini
You can now create custom AI experts called Gems to help with specific tasks or topics (sounds a lot like custom ChatGPTs. Imagen 3, their image generation model, is now incorporated into Geminii as well. They also addes some additional safeguards. Check out their full article here.
The controversial California AI bill that has divided the tech world
California's SB 1047 would require AI developers to comply with certain rules before developing their models. It passed the state Senate 32-1, with the Assembly facing an Aug. 31 deadline to approve the measure in time for it to head to Gov. Gavin Newsom. Elon Musk said Monday that California "should probably pass the SB 1047 AI safety bill," arguing that AI's risk to the public justifies regulation. "This is a tough call and will make some people upset," he said. Read the entire article here. Read the entire bill here.
OpenAI and Anthropic will share their models with the US government
OpenAI and Anthropic have agreed to let the US government access major new AI models before release to help improve their safety. The companies signed memorandums of understanding with the US AI Safety Institute to provide access to the models both before and after their public release. Read the entire article here.
100M Token Context Windows
Magic's recent blog post discusses their development of AI models capable of processing up to 100 million tokens in context, enabling enhanced reasoning and memory during inference. These ultra-long context models can vastly improve tasks like software development by holding extensive codebases and documentation in context. You can read the entire article here.
That's it for this week. Until next week!
Комментарии