Realistic Robotic Wrists, Flying Floaties, and AI News from Microsoft, Google, OpenAI, and Anthropic

Welcome to this week's "Curious About AI" newsletter! I'm Dylan Curious, and I'm excited to share some fascinating developments in the world of artificial intelligence with you.

First up, let's talk about some groundbreaking robotics research. Clone Robotics is designing human-like hydraulic tendon muscles that could revolutionize how we build humanoid robots. Their prototypes show an uncanny ability to mimic human hand movements, gripping objects with impressive dexterity. While still in development, this technology could lead to more lifelike and capable robotic assistants in the future.

In a surprising blend of AI and safety, a new flying drone has been developed that can act as a lifesaver for people who fall off boats. This innovative device could dramatically improve water rescue operations, potentially saving lives in emergency situations. It's a perfect example of how AI and robotics can be used to enhance public safety.

Google's DeepMind, once primarily a research lab, is pivoting towards becoming an AI product factory. This shift, likely in response to competition from OpenAI and Microsoft, highlights the increasing commercialization of AI research. It's a delicate balance between pursuing groundbreaking research and developing practical applications, and I'm curious to see how this change will impact the AI landscape.

Speaking of practical applications, SoftBank has developed an AI that can make angry customers sound calmer on the phone. While this raises some ethical questions, it could potentially reduce stress for call center workers. It's a reminder of how AI is increasingly being used to mediate human interactions.

In the realm of national security, the US Navy is now using AI-equipped underwater drones to scan for threats. These drones use sonar sensors and AI to identify objects on the ocean floor, significantly reducing the time required to clear areas. It's a fascinating example of how AI is being integrated into defense technologies.

On the healthcare front, OpenAI is expanding into the sector with a new partnership focused on cancer care. Their AI assistant, powered by GPT-4, aims to help doctors plan cancer screenings and pre-treatments more efficiently. This development underscores the potential for AI to support medical professionals and improve patient care.

Lastly, I found a clip from an Anthropic employee particularly intriguing. They discussed how their AI model, Claude, is discovering unexpected semantic connections between concepts. For instance, the model seemed to understand the multifaceted nature of veganism across different languages and contexts. This suggests that these AI models are not just pattern-matching, but potentially developing a deeper understanding of complex human concepts.

As always, these developments bring both excitement and important ethical considerations. I encourage you to reflect on how these AI advancements might impact our lives and society as a whole.

Until next time, stay curious!

Warmly, Dylan Curious