Good morning. Today is a big day for the business of artificial intelligence with Microsoft reporting earnings after the bell.

Investors will be super keyed in to Microsoft’s capital expenditure, which last quarter hit a record $14 billion. We’ll have more on this for you tomorrow.

— Ian Krietzberg, Editor-in-Chief, The Deep View

In today’s newsletter:

• 🩻 AI for Good: Identifying stages of breast cancer tumors

• 📱 Apple delays artificial intelligence rollout to IOS

• 🏎️ Watch: AI powers cars in tandem drifting experiment

• 🏅 Google’s Olympics AI commercial highlights a big ethical question

AI for Good: Identifying stages of breast cancer tumors

There is a type of pre-invasive tumor — Ductal carcinoma in situ (DCIS) — that can sometimes progress into deadly forms of breast cancer. But since it is difficult for doctors to determine both the type and stage of DCIS tumors, patients are often overtreated. 

What happened: MIT researchers recently developed a machine-learning model that can identify the different stages of DCIS. 

• The model was trained on a large dataset of breast tissue images; researchers said it could be leveraged by clinicians in the future to streamline breast cancer diagnoses without the need for intensive tests. 

• The model was trained to learn a “representation of the state of each cell in a tissue sample image, which it uses to infer the stage of a patient’s cancer.”

Why it matters: Current tests to determine the stage of DCIS tumors are “too expensive to be performed widely,” according to the researchers. 

The researchers said that this technique is scalable.

Get instant access to NVIDIA H100, L40S and V100 GPUs

Get immediate access to up to 8 NVIDIA® GPUs, along with CPU resources, storage, and additional services through our user-friendly self-service console.

Learn more here today!

Apple delays artificial intelligence rollout to IOS

Bloomberg’s Mark Gurman reported Sunday that Apple is delaying its genAI rollout — known also as Apple Intelligence — a few weeks to October. 

The details: Bloomberg reported, according to anonymous sourcing, that Apple Intelligence will miss Apple’s big September IOS overhaul. It will instead roll out a few weeks later, to give the company time to fix bugs. 

• Apple did, however, release the first beta version for IOS 18.1 (which includes a beta version of Apple Intelligence) to developers on Monday.  

• When Apple Intelligence does launch, it won’t be launching all at once; Bloomberg reported that some of the biggest features — including the Siri overhaul — will be incorporated via additional software updates after the initial rollout. 

Apple did not respond to a request for comment. 

Apple is among four Magnificent Seven tech giants reporting second-quarter earnings results this week. 

• Legal AI startup Harvey raised $100 million in a Series C funding round.

• Dental AI startup Pearl raised $58 million in a Series B funding round.

• Websites are Blocking the Wrong AI Scrapers (Because AI Companies Keep Making New Ones) (404 Media).

• Elon Musk posts deepfake of Kamala Harris that violates X policy (The Verge).

• Hewlett Packard set for unconditional EU nod for $14 billion Juniper deal, sources say (Reuters).

• AT&T, other phone companies sued over stolen nude images could face liability after court ruling (CNBC).

• At the Olympics, AI is watching you (Wired).

Watch: AI powers cars in tandem drifting experiment 

A Stanford research team, in collaboration with Toyota Research Institute, has created the first autonomous tandem drifting team. Their goal is to apply what they learn on the track to make all vehicles safer. 

But first, what is tandem drifting? Tandem drafting is a sport that’s kind of like pairs figure skating but in cars. The lead car drifts around a track, while a follow car has to mimic the lead car’s movements.

The details: In this scenario, both the lead car and the follow car — sometimes separated by no more than 10 inches — were driven, not by people, but by AI. 

• The researchers developed a neural network vehicle model to accomplish the task. 

• The researchers called tandem drifting “perhaps the ultimate expression of car control capability.” The cars have to make constant, minute adjustments to the steering, gas and brakes, since a “slight wrong move can cause a vehicle to spin out.”

Why it matters: “The things that we’ve been developing here can one day make the roads much safer for everybody,” Trey Weber, one of the engineers, said, referring to advanced driver-assist technology designed for help on slick, slippery roads.

Invest smarter with Public and AI

If you're looking to leverage AI in your investment strategy, you need to check out Public.

The all-in-one investing platform allows you to build a portfolio of stocks, options, bonds, crypto and more, all while incorporating the latest AI technology — for high-powered performance analysis — to help you achieve your investment goals.

Join Public, and build your primary portfolio with AI-powered insights and analysis.

Google’s Olympics AI commercial highlights a big ethical question

In case you missed it, Google’s been running an ad for Gemini during the Olympics. 

The first half of the ad is super inspiring; a runner’s daughter is regularly training to become a track star. Her inspiration: the real-life Olympic track star Sydney McLaughlin-Levrone. 

But the second half of the ad starts like this: “She wants to show Sydney some love, and I’m pretty good with words, but this has to be just right. So Gemini, help my daughter write a letter telling Sydney how inspiring she is. And be sure to mention that my daughter plans on breaking her world record one day. She says ‘sorry, not sorry.’” 

We never see the drafted letter that Gemini generates in response to that prompt. But the idea is simple: “a little help from Gemini.” 

I tried out this prompt on Gemini, and here’s what I got: 

The idea of this, according to Shelly Palmer, a professor of advanced media at Syracuse University, is “disturbing.” 

The problem: Language, Palmer said, provides the framework for society’s shared reality. Think about it — different languages parse up the world in different ways. For example, ‘blue’ does not exist the way English speakers might expect in the Russian language; there are instead two separate words, one for light blue and one for dark blue. 

• There are plenty of similar examples, from the way we give directions, to the way past, present and future are described. 

• “Google would have us believe that this young girl doesn’t need to learn to articulate and describe her reality. This is criminally negligent,” Palmer said. 

If everyone chose to communicate through chatbots in this manner, he said, “it is easy to imagine a future where the richness of human language and culture erode.”

This sends me back to an older question relevant to generative AI: what problem does it solve, and what is the cost-benefit analysis of the solution? 

In this instance, it seems that the ‘problem’ is emotive human communication (so, not really a problem). The cost of this solution, meanwhile, is plenty of outsized carbon emissions from an already over-stressed electrical grid. 

This is not the AI that has been promised. This is not a cure for cancer. This is not a silver bullet for climate change. 

This seeks to automate creative processes, rather than mundane processes. It seeks to cast the highest function of the human spirit as somehow out of reach of humanity. The dad can’t get the words “right,” but AI can. 

• The idea of that feels especially twisted, as the idea of ‘correct’ doesn’t — and shouldn’t — exist in emotive communication and human creativity. 

• And the idea of an eight-year-old handwriting a note — full of grammatical errors and misspellings — to Sydney,  rather than a genAI system predicting what a kid might think, feel or want to say, feels infinitely more powerful and tremendously more real than the cold, emotionless probabilities that this ad conveys. 

I worry about a world that is taught to process its own communication through an unthinking, unfeeling machine. I worry what will be lost. 

As Jane Rosenzweig, the director of the Harvard Writing Center, wrote: “the ad suggests that we outsource what really matters so that it will be ‘perfect,’ which seems like a way to make what really matters … matter less. Which is why I worry about outsourcing first drafts more generally, since that's where we're figuring things out.”

There is a way they could have done this that wouldn’t taste so bitter; a translation, perhaps, of a letter written in an obscure language. Literally, a “little” help from Gemini. I think the avenue Google chose to demonstrate how Gemini can be used is telling. 

A poll before you go

Thanks for reading today’s edition of The Deep View!

We’ll see you in the next one.

Your view on whether Apple can be trusted to develop AI responsibly:

Around a third of you believe that no Big Tech company can be trusted; around 20% of you said that Apple’s specific agreement with the White House still doesn’t make them trustworthy.

Only 15% of you trust Apple to develop AI responsibly following its agreement with the White House.

I trust Apple:

• “At the moment Apple’s fortune is soft/hardware smooth smart disruptive integration and mass market of cool devices, not pumping/selling their users data. It can change in the future but actually I trust this company when Apple´s CEO talks about respecting customers’ privacy.”

^{*Public disclosure: All investing involves the risk of loss, including loss of principal. Brokerage services for US-listed, registered securities, options and bonds in a self-directed account are offered by Public Investing, Inc., member FINRA & SIPC. Cryptocurrency trading services are offered by Bakkt Crypto Solutions, LLC (NMLS ID 1828849), which is licensed to engage in virtual currency business activity by the NYSDFS. Cryptocurrency is highly speculative, involves a high degree of risk, and has the potential for loss of the entire amount of an investment. Cryptocurrency holdings are not protected by the FDIC or SIPC.}

^{Alpha is an experiment brought to you by Public Holdings, Inc. (“Public”). Alpha is an AI research tool powered by GPT-4, a generative large language model. Alpha is experimental technology and may give inaccurate or inappropriate responses. Output from Alpha should not be construed as investment research or recommendations, and should not serve as the basis for any investment decision. All Alpha output is provided “as is.” Public makes no representations or warranties with respect to the accuracy, completeness, quality, timeliness, or any other characteristic of such output. Your use of Alpha output is at your sole risk. Please independently evaluate and verify the accuracy of any such output for your own use case.}