In this episode we cover the broad releases highlighted by Google in their I/O event, a partnership between OpenAI and Reddit, as well as the release of the multi-modal GPT-4o model. All in just three minutes.

In this episode we cover a concerning pre-print which shows that GPT-4 can exploit 87% of one-day vulnerabilities, a stealthy release of "gpt2-chatbot" that we know nothing about, and Danti's new AI that allows analysts to search satellite imagery with natural language. All in just three minutes.

In this second deep dive episode we cover the history of AI as a foundation for our future deep dives in modern AI approaches.

In this episode we cover Rio (an AI-driven news anchor), Sanctuary AI's newest Fenix humanoid robot, and the incredible developments in AI generated music, all in just 3 minutes!

In this episode we cover Meta's Llama 3 release, Boston Dynamic's new Atlas robot, and evaluation criteria for AI in medicine, in just three minutes.

In our first deep dive, we look at the history of computing. These deep dives will be less frequent than our regular three stories in three minutes, but will serve as a primer to dig into more rich content and explore concepts that take more than 3 minutes.

In this episode we discuss the new AI Pin from Humane, the release of Llama 3 from Meta at a reported size of 140 billion parameters, and Gemini 1.5 Pro with a context window of 1 million tokens, all in just three minutes

In this episode we cover SiMa.ai's newest multimodal chipset which could allow you to run AI on your devices, Tesla's "human-like" update to full self-driving, and Meta's approach to label (rather than remove) AI generated content.

In this episode we discuss the first ever patient to receive a Neuralink implant, an automated code-scanning autofix for cybersecurity, and a $4 billion investment in AI by Amazon