People Are Talking

From Protocol: Investors are pouring money into quantum computing, but Sequoia's Bill Coughran plans to wait before committing to the software side:

• "I think the challenge for them is some of the abstractions and the way that actual hardware work hasn't been completely sorted out, so there's some risk that they'll go write a lot of software that doesn't necessarily map onto the hardware that gets developed."

Representative Pramila Jayapal said she tried to work with Amazon privately on her concerns about the company, but now is going public:

• "Two things can be true at the same time. A company can be doing tremendous work that is incredibly valued and essential, and it can be treating workers badly."

The highest-flying startups could be the ones set to fall the farthest, Keith Rabois warned:

• "There's a correction going on underneath the hood in many of these companies completely independent of a virus. The virus just put a spotlight on it and to some extent convinced some nonbelievers."

Eric Schmidt is trying to push the Pentagon to move as fast as Silicon Valley, and it's a struggle:

• "I am bizarrely told by my military friends that they have moved incredibly fast, showing you the difference of time frames between the world I live in and the world they live in."

The Big Story

Can regulators keep up with tech keeping up with the virus?

Passing a broad, national data-privacy bill has proven essentially impossible. But with a pandemic comes certain political openings, and one group of senators is determined to make the most of it.

• As Protocol's Emily Birnbaum reported a few days ago, a group of Senate Republicans led by Roger Wicker will introduce a bill as soon as this week called the COVID-19 Consumer Data Protection Act. It seeks to put guardrails on the data that companies can collect in the name of virus-tracking, and punish companies that break the rules.
• The bill applies to any project that uses data to track the spread of the virus, and would prevent companies from repurposing the data they collect for other uses.
• Wicker also led a call for "enlisting big data in the fight against coronavirus" in early April, with a statement that seemed … dramatically less worried about the privacy risks. Things move fast, I guess?

The CCDPA seems to have a decent shot, especially if it's rolled into the next aid bill.

But the big question mark is enforcement. Sure, Apple and Google have promised to shut down their remarkably powerful Bluetooth-tracking systems when this is all over, but who's going to make sure?

• One privacy counseltold The Verge that the bill doesn't create a new rule-making authority, or give the FTC or anyone else resources to enforce the rules. The bill would also supersede any state laws on the subject, blocking the path for more serious legislation.

Similar debates are happening all over the world, too:

• A government-endorsed app in India has already been downloaded more than 75 million times, and has raised big privacy concerns.
• The U.K. is testing a few ideas, including an "immunity passport" and a location-based app that's being tested on the Isle of Wight.

There's still no consensus for contact tracing, either for tech or for data security. But if we're ever going to get a universally agreed-upon approach, I'd bet on the one from Google and Apple.

• The CDC, for instance, published new contact-tracing guidelines last week, which match neatly with what Google and Apple are working on. The two companies are releasing much more information on how their tech can be used this week, too — stay tuned.

Meantime, there's also this piece from the Brookings Institute, which makes a compelling argument for why the whole idea of contact tracing won't get us anywhere.

Apps

Coronavirus apps aren't catching on yet

All this brings me back to that pesky problem with contact tracing: It only works if everybody uses it. Or at least almost everybody: The U.K. government is aiming to get 80% of smartphone users, which amounts to 56% of the country's population or about 37 million people, participating in contact tracing. As I wrote last week, U.S. polls suggest that getting to that number would be a challenge.

How successful are existing virus-related apps that are out there? They're not doing great:

• Johns Hopkins released an app on Friday called COVID Control that Sensor Tower's data says has been downloaded fewer than 5,000 times.
• HowWeFeel, an app created by Pinterest CEO Ben Silbermann, peaked at #44 on the App Store in early April, but is now down to #1393. The app appears to have about 485,000 users, which is a lot! But it's nothing like a meaningful global network.
• The COVID Symptom Tracker app has been a hit in the U.K., where its creator said it has more than 3.1 million users. It's also available in the U.S. but hasn't made much of a splash here.
• Apple's own screening app had a big debut, hitting #4 on the App Store. But it dropped fast, down as far as #960 on the charts.

One way companies are trying to make things easier is by building web apps, so users can click a link instead of downloading an app. And when Google and Apple build the infrastructure straight into their phones, it could lower the barrier to entry, at least for those who want in. We'll have to wait and see if that's enough.

A MESSAGE FROM WORKDAY

The Workforce of Tomorrow Requires Better Tools Today

The role for government centers on deriving better data sets, enabling better credential interoperability, and creating better reskilling incentives.

Read more here

Machine Learning

Computers learn to peek behind the face mask

Speaking of barriers to entry: Perhaps the most annoying side effect of wearing a mask every day is that my Face ID doesn't work anymore. Luckily Apple's working on it, but it turns out mask-face-recognition is a pretty interesting and important challenge for a lot of people.

• A few organizations, including U.S. Customs and Border Protection, told Wired they're able to identify people even with masks on. Developers in China have been working on this for a while — with apparently mixed results.
• But as companies develop their recognition programs, many are struggling with how to deal with anything other than a head-on, well-lit face. Which is to say, they're struggling with how to deal with the real world.

As we go back to touchless offices, try to stop handing over our credit cards and in general learn to keep our grubby mitts to ourselves, there are going to be plenty of new places where facial recognition could work. Whether it can see through the cool cloth mask I bought on Etsy will be a telling sign of just how helpful technology will be in helping us return to normal.

Coming Up This Week

Two Protocol events this Thursday! At noon PDT / 3 p.m. EDT, we'll have our latest Virtual Meetup, with Biz Carson and some of the best VCs in tech talking about what's happening now — and where we go next. Sign up here.

Then, at 5:30 p.m. PDT / 8:30 p.m. EDT, we're co-hosting an event with sf.citi and sea.citi, all about how tech has responded to the pandemic. We'll be chatting with executives from Twitter, Slack, Postmates, and many other companies. Sign up here.

We're in week three of Coronavirus Earnings Season: Pinterest, Uber, Lyft, Shopify, Roku, Nintendo, Cloudflare, Dropbox and more all report this week. Keep track of them on our earnings guide.

IBM's Think Digital conference starts tomorrow — it'll be an early chance for new CEO Arvind Krishna to talk about the future of the company.

In Other News

• Today in coronavirus: Google built an AI tool for helping banks quickly review and assess PPP loan applications. Anti-lockdown protestors, forced off Facebook, are congregating on MeWe instead. Uber will soon start requiring drivers and passengers to wear masks in the car, and is developing tech to ensure that happens. Twitch streamers are striking in solidarity with gig workers. And while coronavirus strikes aren't managing to stop business, they're still being called a success.
• Intel reportedly bought Moovit, an AI-navigation startup, for about $1 billion. The company will try to help Intel solve traffic and routing and many more seemingly small but actually incredibly complex things that stand in the way of self-driving.
• Tesla applied to be an electricity supplier in the U.K. It doesn't seem to want to be a power company, per se, but might instead be looking to take its Autobidder energy trading platform to the country.
• You know that company Eko? The one that's suing Quibi for stealing its tech? Turns out the lawsuit is reportedly being funded by Elliott Management, which also took a stake in Eko as part of the deal to finance the legal challenge.
• Salesforce built a new set of features called Work.com, designed to help businesses of all sorts get back to work. It includes everything from a contact-tracing tool to workplace reskilling systems, all of which will be out in June.
• Don't miss this story from The Washington Post about how Bill Gates became one of the leading experts on the pandemic — a threat he's been warning about for years.
• After Facebook invested in Jio, Silver Lake jumped in too. It's putting $747 million into the Indian giant, at a $65 billion valuation that makes Jio one of the world's largest startups.
• How will the NBA get back to work? It has a complicated, expensive, potentially impossible plan — that offers inspiration for anybody else who is trying to reopen an office, restart a conference, or just figure out how to keep people safe while things get back to normal.

One More Thing

You get a meme. And you get a meme!

The best thing on Twitter the last couple of weekends was the memes. OK, that's always true, but these were particularly good memes — they were made by a neural network called This Meme Does Not Exist, trained on 48 popular meme templates and a huge library of human-made memes. (The technical details are pretty interesting, too.) The resulting greatest-hits AI Meme Stream is an incredible mix of unintentional comedy and hysterical nonsense. Pro tip: The Galaxy Brain template is a gold mine.

A MESSAGE FROM WORKDAY

The Workforce of Tomorrow Requires Better Tools Today

The role for government centers on deriving better data sets, enabling better credential interoperability, and creating better reskilling incentives.

Read more here