Skip to Content

Data Sheet—Intel’s A.I. Plan Vs. Donald Trump’s

A month ago Donald Trump issued an executive order designed to maintain America’s greatness on artificial intelligence. I reviewed it unfavorably for its lack of specifics, absence of funding details, and preponderance of management-consultant fluff.

About a month later the semiconductor giant Intel offered its own A.I. plan, a 13-page whitepaper of its recommendations for a U.S. strategy on A.I. Intel cites the work of many management consultants, and it uses (in an accompanying fact sheet) some interestingly diplomatic language, like grouping together China, India, Japan, and the European Union as “global neighbors” to the U.S. (One president’s punching bags are a multinational corporation’s neighbors.)

On balance, though, the Intel proposal is a cut above the White House’s. It too refrains from offering specific dollar values, but Intel’s report does specify where money should be spent. It calls, for example, for government funding on research to determine the best areas for A.I. spending and for allocations for fundamental research, much as Washington spends heavily on health research.

Intel has much to gain from a national A.I. strategy; its chips should power the machines and sensors that A.I. enables. But offering one also places the company in a precarious position given its significant global operations, particularly in China.

That a company as serious and substantial as Intel feels compelled to offer a “national strategy” on A.I. shows that this topic is more than just hype—though there’s plenty of hype around it too. To that end, Fortune is pleased to announced its newest newsletter, “Eye on A.I.,” a weekly compendium of business-related A.I. articles with a dash of considered viewpoints and original reporting. As readers of Data Sheet, we know you are interested in technology. That almost by definition means you want and need to know more about A.I., and I invite you to sign up for the newsletter here.

Below you’ll find a sampling of the type of material you can expect to find in Eye on A.I., which launches March 26. My colleague Jonathan Vanian, who follows enterprise technology for Fortune, will curate the newsletter. We hope you enjoy it. Maybe the White House will subscribe. (I also highly recommend “Finding New Cures in Old Drugs,” a feature in the issue of Fortune about two companies that are deploying A.I. to ‘map’ the complicated pathways through which neurodegenerative diseases like Parkinson’s ravage the body.)

Subscribe to Eye On A.I.

Adam Lashinsky


Siri may get smarter. Apple has paid an undisclosed amount for startup Laserlike, which created a machine learning app that keeps people up to date on news and articles they’re interested in. Tech news site The Information reported that Laserlike, founded by ex-Google engineers, will add muscle to Apple’s A.I. group and could help improve Apple’s Siri digital assistant.

China to surpass the U.S. China will likely produce more high-quality A.I. research papers than the U.S. by 2020, according to the Allen Institute for Artificial Intelligence. The non-profit’s analysis of more than two million academic A.I. papers “shows a clear trend of Chinese ascendance in the field of A.I.”

Spending big bucks on data labeling. Australian data technology company Appen plans to buy startup Figure Eight for nearly $300 million, Appen said. Figure Eight, whose investors include Salesforce Ventures and Trinity Ventures, helps companies label and annotate their data for deep learning projects.

Want free A.I. hardware blueprints? Facebook has released free designs for a new server and specialized computer chips that are intended for tasks like speeding up data training for deep learning projects. The social networking giant typically open sources its data center hardware blueprints so that other companies can inspect and improve the designs.

Stanford gets A.I. organized. The university announced it has formed the Institute for Human-Centered Artificial Intelligence to be run by professor of computer science and former director of the Stanford AI Lab Fei-Fei Li and professor of philosophy John Etchemendy. The aim is to create an interdisciplinary, global hub for A.I. research in fields of study ranging from computer science and robotics to business, law, and medicine.


Tell the truth. U.K.-based venture capital firm MMC Ventures analyzed 2,830 European “A.I.” startups and discovered that 40% of the companies don’t actually use machine learning in their products but instead use conventional data analytics technology.

Lower your expectations. Be skeptical of cybersecurity companies that promise A.I. that can “automatically detect unknown attacks” in corporate networks, according to trade publication Information-Age. Machine learning is good at helping security staff sift through lots of data in corporate networks to determine “signals” that could indicate a hack. But the article said that technology won’t replace a company’s security teams, who have the expertise to interpret the data.


Morgan Stanley A.I. researcher Sasha Luccioni has left the bank’s machine learning team in Montreal to become a postdoctoral researcher at The Mila-Quebec Artificial Intelligence Institute, founded by deep learning pioneer Yoshua Bengio.

Hearst Magazines named Mike Smith as the company’s chief data officer, a newly created role leading the company’s digital advertising and data operations. Smith was previously a vice president in Hearst’s digital unit in charge of revenue platforms and operations.

Boston’s mayor appointed Stefanie Costa Leabo as the city’s chief data officer, overseeing various data-related projects. She was previously Boston’s director of performance management.

Credit bureau Equifax added data executive Heather Wilson to its board. Wilson is the chief data scientist of fashion retailer L Brands.


Deep Learning as a solar energy tool. Duke University researchers published a paper on how deep learning could be used to predict the energy capabilities of solar panels scattered across the U.S. The researchers trained an A.I. system on 16,000 labeled images of solar arrays in California, and then used the system to discover and map solar arrays and their projected energy outputs across Connecticut.

Deep Learning for safeguarding wireless networks. Researchers at Google’s DeepMind unit published a paper describing the use of deep learning as a cybersecurity tool to protect wireless networks from hacks. The researchers’ deep learning techniques are intended to make it difficult for hackers using similar A.I. techniques to guess certain attributes in wireless networks that could make them vulnerable.

Deep Learning to predict the risk of going blind. Researchers from Stanford University, University of Illinois, and medical tech company Carl Zeiss Meditec published a paper detailing how deep learning can be used to predict the risk of people developing age-related macular degeneration, one of the leading causes of blindness. The A.I. techniques could help doctors track the progression of the “dry” version of macular degeneration to the more severe “wet” version.


Darktrace CEO: The Future of Cybersecurity is A.I. vs. A.I. – By Robert Hackett

Millions of Flickr Photos Were Scraped to Train Facial Recognition Software – By Emily Price

Google’s A.I. Assistant Wants to Make Restaurant Reservations For You – By Alyssa Newcomb

A Record Number of Robots Were Put to Work in the U.S. in 2018 – By Natasha Bach

Facebook Is Using A.I. to Stop Revenge Porn – By Emily Price


Experts disagree on how A.I. will change our lives. Two recently published books based on interviews and essays from the world’s A.I. pioneers are frustrating because the experts often disagree about some of A.I.’s most important questions. Vox reviewed the books Possible Minds and Architects of Intelligence and found that they make for “gripping reading,” but they don’t provide a simple conclusion about how A.I. will transform society over the next few decades.

Among the disagreements:

  • Whether A.I. will experience another “winter” like during the early 1990s.
  • Whether neural networks will lead to more computer science breakthroughs or are just a fad.
  • When computers will truly reach a level of intelligence akin to human beings.

As Vox’s Kelsey Piper wrote, “The disagreements on display in these anthologies aren’t just charming intellectual spats — they’re essential to the policy decisions that we need to make today.”