Collaborative Intelligence Week in Review - 10May2024
Best business article(s) I read this week…
The Art of Asking Smarter Questions
This Harvard Business Review article isn’t directly related to AI; it discusses ways to improve strategic decision-making. Prospering in the age of AI will require leaders to make many different types of strategic decisions, so I think this is appropriate.
Best technical article(s) I read this week…
The authors explore the challenge of pairing computationally powerful intelligent agents with less powerful ones (such as humans). In the paper's example, a chess engine can suggest a move that a human follows, but because the human cannot compute all possible outcomes on subsequent moves, the human cannot select an appropriate follow-on move without the aid of the chess engine. As the power and capability of collaborative technologies continue to grow rapidly, this research has practical applications in most problem domains. How do we best pair humans with powerful artificial intelligence in ways that the two can work together to achieve the desired outcomes?
This is not particularly technical, but it doesn’t qualify as a business article. It’s from MIT Technology Review in association with Jina AI. Multimodal AI uses information in different “modalities” (e.g., text, images, sounds) to create a more nuanced view of the world—one that is more akin to human perception.
News items I found interesting…
Siemens, Microsoft Working to Standardize Digital Twin Definition Language
The two companies are working with the W3C Consortium to combine a newly emerging Digital Twin Description Language with W3C’s existing “Thing Description Language.” The goal of the work is to make it easier to share digital twin models and related data.
Generative AI used to improve robots’ movements
Researchers at Tohoku University in Sendai, Japan, have used deep reinforcement learning to understand the neural circuit patterns in the human spinal cord and apply them to robotics, resulting in an improvement in robots’ ability to simulate human walking and running.
Google DeepMind’s AlphaFold 3 can predict the interactions of “all of life’s molecules”
The company’s new model improves prediction accuracy by over 50%, which should enable much more rapid drug discovery and more effective drug treatments.
AI and Healthcare Simulation: The Shifting Landscape of Medical Education
This article describes how AI can be used to create realistic simulations for use in medical training. While specific to medicine, the general approach can be applied to any industry or specialty.
Coolest thing I saw…
Penguin-inspired robot explores the sea using AI
This autonomous underwater vehicle was 3D printed, and it uses AI to recognize objects and navigate. It moves through the water by mimicking the movement of penguins
A company that caught my eye…
Despite its pedestrian name, this Abu Dhabi-based company is involved in AI, genomics, and spacecraft. This Bloomberg BusinessWeek profile on the company presents a lot of interesting facets, including the geopolitical struggle over AI dominance between the US, China, and other state actors.