In a long-awaited move, U.S. President Joe Biden's recent Executive Order on Artificial Intelligence is viewed by many as a substantial step towards regulating AI, especially considering the U.S. is the cradle to numerous influential companies pushing AI boundaries. The order, as experts like Lee Tiedrich of Duke University's Initiative for Science & Society commend, is a "creative" bundle of initiatives leveraging the executive branch's reach, amid the limitations of not being able to enact legislation or directly set rules.

The Executive Order builds upon previous efforts including the "Blueprint for an AI Bill of Rights" and voluntary commitments from 15 leading AI companies towards managing AI risks. This initiative occurs parallelly with significant regulatory strides globally, such as the European Union's impending AI Act and China's swiftly enacted laws on AI recommender systems and generative AI.

The order, though detailed, is outlined in a fact sheet with the full text expected soon. It discusses a range of topics from safety and security to civil rights. For instance, the National Institute of Standards and Technology (NIST) is tasked with establishing rigorous standards for red-team testing to ensure safety pre-public release of AI systems. It also addresses the need to curb algorithmic bias, which resonates with Brown University's Suresh Venkatasubramanian who appreciates the order's attempt but hopes for stronger regulation in law enforcement use of AI.

Privacy concerns, especially around biometric data, are echoed by Cynthia Rudin, a Duke University professor. While appreciating the order's ambition, she wishes for a clearer stance on biometric technologies. The order's call for Congress to enact bipartisan data privacy legislation also aligns with ongoing efforts by Senator Chuck Schumer, although its incorporation remains to be seen.

A notable aspect of the order is the directive for content authentication and watermarking to label AI-generated content, addressing the burgeoning issue of synthetic media. However, the practicality of watermarking, especially on text-based deepfakes, remains a skeptical point for Rudin.

On a broader spectrum, Susan Ariel Aaronson, a professor at George Washington University, praises the order but hopes for a more structured approach towards governing AI, highlighting the current "patchwork of principles, rules, and standards."

The Executive Order hints at a decentralized approach to AI governance, spreading responsibilities across federal agencies, each overseeing AI in their respective domains, a move appreciated by experts like Tiedrich.

This order precedes the UK's AI Safety Summit, where Vice President Kamala Harris will represent the U.S., emphasizing the nation's renewed stance on AI governance. Amid global endeavors to regulate AI, the U.S.'s latest move is a significant stride, albeit with experts and stakeholders keenly awaiting the detailed text and its subsequent implementation in a rapidly evolving AI landscape.

Apple has released new Macs powered by M3 chips, showing they are embracing AI through custom hardware. The M3 Max chip targets AI/ML developers with its powerful GPU stack.

Apple’s news M3 chips have new GPUs optimized for AI, up to 2.5x faster than M1.

What does that mean?

The new M3 chips have a new GPU architecture with a feature Apple calls dynamic caching. Dynamic caching allocates GPU memory adaptively in real-time, increasing the utilization of the GPU. The new neural engine is 60% faster, speeding up ML.

The M3 Max has a 16-core CPU and 40-core GPU supporting 128GB of unified memory. Apple claims that’ll allow developers to work with models with billions of parameters.

Apple is keeping pace with AI-focused offerings from rivals like Qualcomm. Qualcomm made a similar claim this month with their Snapdragon X Elite of being able to run a 13B model on-device.

Why you should care?

If you use graphics/ML apps, the new Macs will bring meaningful speed improvements. If you’re a developer, Apple's platform is now more compelling for generative AI projects needing local compute.

Genie AI: Your AI legal assistant. Understand, customize and negotiate any legal document.

timeOS: Record, transcribe, and summarize your Slack Huddles. 

Parsio : A tool to analyze your PDFs, e-mails and other documents. The tool also helps you extract the best information.