Future Vision: From Shades to Predictive Health

Hey Ai5ers,

This week’s AI5 newsletter is packed with exciting AI breakthroughs, from Meta’s latest AI powered smart glasses, to AI tools that can predict diseases decades in advance, to Nvidia’s bold play in chipmaking.

Let's dive in!

Today's Ai5:

👀 Meta's newest AI powered smart glasses
🤔 Which diseases will you have in 20 years?
💰 Nvidia buys $5B stake in Intel, planning AI chip collaboration
👎 China tells its tech companies they can't buy AI chips from Nvidia
☝️ Silicon Valley bets big on "enviroments" to train AI agents

Prompt of the Day 🎨

September 19th is 'Talk Like A Pirate Day!,' the prompt:

"Create a photo-realistic poster promoting 'Talk Like a Pirate Day.' Show men and women dressed as pirates in a lively setting, such as a dockside tavern or ship deck. Include authentic pirate details like tricorn hats, bandanas, eye patches, swords, and tankards of rum. The scene should feel festive and adventurous with dramatic lighting. Overlay bold headline text that reads: 'Talk Like a Pirate Day – September 19!' Add pirate phrases in weathered text or speech bubbles, such as 'Ahoy Matey!' 'Yo Ho Ho!' 'Shiver Me Timbers!' 'Avast, Ye!' The overall design should look like an advertisement poster with a swashbuckling theme."

Meta's newest AI-powered smart glasses 👀

(Courtesy Meta)

Meta CEO Mark Zuckerberg took the stage on Wednesday to unveil the company’s next-generation artificial intelligence-powered wearable device: a pair of smart glasses with a tiny display inside the lens.

The Displays feature a tiny display screen on the inside right corner of the right lens. The display looks as though it’s projected several feet in front of the user’s surrounding environment.

That display makes it possible to do a variety of things one might previously have done on their phone screen: view and send messages, capture and review photos and videos, watch Instagram Reels and take video calls where users will see the person on the other end.

There’s also a navigation feature that shows where a user is on a map in real-time, so they could walk somewhere without staring at a maps app on their phone. And live captioning and translation lets users see what a conversation partner is saying in real-time.

Full article by Clare Duffy

Which diseases will you have in 20 years? 🤔

A new artificial intelligence (AI) tool can forecast a person’s risk of developing more than 1,000 diseases, in some cases providing a prediction decades in advance.

The model, called Delphi-2M, uses health records and lifestyle factors to estimate the likelihood that a person will develop diseases such as cancer, skin diseases and immune conditions up to 20 years ahead of time. Although Delphi-2M was trained only on one data set from the United Kingdom, its multi-disease modeling could one day help clinicians to identify high-risk people, allowing for the early roll-out of preventive measures.

The researchers trained Delphi-2M on data from 400,000 participants of the UK Biobank, a long-term biomedical monitoring study.

Full Article by Gemma Conroy

Nvidia buys $5B stake in Intel, planning AI chip collaboration 💰

Nvidia has agreed to buy a $5 billion stake in Intel as part of a broader deal to together develop “multiple generations” of data center and PC products.

The companies will integrate their two architectures using Nvidia’s NVLink interface, which enables data and control code transfers between CPUs and GPUs. NVLink enables faster transfers between chips compared with other standards like PCI Express; that’s crucial for AI applications, which require many GPUs to run together and process immense workloads.

For data centers, Intel will manufacture a new line of x86 CPUs specifically customized for Nvidia’s AI infrastructure platforms, to be offered to enterprise and hyperscale customers.

Full article by Ram Iyer

China tells its tech companies they can't buy AI chips from Nvidia 👎

Nvidia just got shut out of the Chinese market — this time by the Chinese government instead of the US.

The agency also told tech companies including ByteDance and Alibaba to stop testing and ordering Nvidia’s RTX Pro 6000D server, a device designed specifically for the market in China.

This ban will deliver quite a blow to China’s tech ecosystem. While companies like Huawei and Alibaba design AI chips locally, Nvidia is by far the global market leader, and its chips are considered to be some of the most advanced on the market.

In July, the Trump administration reversed course and gave semiconductor companies the green light to sell their chips in China again. In August, the White House announced it would grant the licenses needed to sell in China, but with a catch: the U.S. government would get 15% of the revenue from the chips sold. But as of Nvidia’s latest earnings, the company had yet to sell any units to Chinese customers under the plan, citing the slow implementation of President Trump’s proposal.

Full article by Rebecca Szkutak

Silicon Valley bets big on 'environments' to train AI agents ☝️

For years, Big Tech CEOs have touted visions of AI agents that can autonomously use software applications to complete tasks for people. But take today’s consumer AI agents out for a spin, whether it’s OpenAI’s ChatGPT Agent or Perplexity’s Comet, and you’ll quickly realize how limited the technology still is. Making AI agents more robust may take a new set of techniques that the industry is still discovering.

One of those techniques is carefully simulating workspaces where agents can be trained on multi-step tasks — known as reinforcement learning (RL) environments.

At their core, RL environments are training grounds that simulate what an AI agent would be doing in a real software application. One founder described building them in recent interview “like creating a very boring video game.”

For example, an environment could simulate a Chrome browser and task an AI agent with purchasing a pair of socks on Amazon.

Full article by Maxwell Zeff