What SoftBank really wants
Some in Silicon Valley aren’t quite sure what to make of SoftBank and its massive, roughly $100 billion Vision Fund. At times, they say privately, it looks like a drunken gunslinger, firing off massive checks in quick succession.
But sources close to SoftBank say there is a method to its madness. In fact, these same sources say SoftBank’s investors believe they’ll see at least a 20 percent internal rate of return (IRR) over time from Vision Fund as it funds whole sectors being disrupted by artificial intelligence and machine learning — from pharma, to utilities, to ridesharing — and whose data SoftBank can leverage into an endless stream of opportunities.
The idea, these people say, is not to produce venture-like returns. The idea is instead to return more money to investors than private equity firms like KKR, whose first 18 private equity funds wound up delivering more than two times total capital invested on a gross basis, and produced a net IRR of 18.9 percent. Says one source close to SoftBank, “If someone is investing in [Vision Fund], he’s expecting to get better returns than with KKR and Blackstone.”
Indeed, 20 percent IRR over seven years — the time SoftBank estimates it will take most of Vision Fund’s bets to play out — is the “worst-case scenario” says one source. “Best case,” adds this person, is “investors get close to what Masa has done in the past.”
It’s a reference to the 44 percent IRR on investments that SoftBank can boast over its 18-year history, though more than one critic has noted that much of this number is rooted in SoftBank founder Masayoshi Son’s early bet on Alibaba, beginning in 2000. Son would eventually pour $58 million into the company; those holdings, which SoftBank maintains, save for a $10 billion chunk it sold to finance another purchase, are currently worth $130 billion.
Higher and higher
Doing back-of-the-napkin math on this 20 percent IRR — whether over seven years or a more traditional 10-year time frame — would translate into between $130 billion and $430 billion for SoftBank’s investors — minus its initial investments, management fees and the debt that makes up roughly $44 billion of Vision Fund’s total holdings.
That’s a whole lot of capital to generate for limited partners, so how does it do it? SoftBank thinks it can get there largely through ridesharing, say sources familiar with its thinking. More specifically, SoftBank is counting on the smooth evolution of today’s rideshare companies into vast networks of self-driving taxis.
It has already made an array of bets that underscore this theme, including on the China-based ride-hail giant Didi Chuxing; in Grab, the dominant ride-hail startup in Southeast Asia; and in Ola,India’s leading ride-hailing company, which reportedly closed on $2 billion just yesterday, including from SoftBank.
Helping grow a U.S. player is also crucial to its strategy, and SoftBank has been openly unsentimental about whether that means funding Uber or Lyft, though Uber would seem to be its strong preference. Says one source close of a meeting that’s slated to take place today, wherein Uber’s directors will vote on whether to go forward with a $10 billion stock sale to SoftBank: “Uber should be scared of SoftBank funding Lyft. They better take [the money].”
We’ll see soon enough how scared or not Uber may be of scorning SoftBank. Certainly, though, concern about the companies that SoftBank doesn’t fund is growing in Silicon Valley.
Asked onstage earlier this month about SoftBank’s impact on Silicon Valley, venture capitalist Steve Jurvetson of DFJ called SoftBank a “kingmaker of sorts that’s giving a massive infusion to some companies and not others.
“In the long run,” Jurvetson said, “that’s just noise. The better product and service should win. In the short run,” he continued, “it could [create] some interesting shifts in the outcomes of companies that would otherwise be in a normal horse race.”
Another source who invested recently in the two-year-old, indoor farming company Bowery Farming, said he was taken aback when SoftBank led a $200 million investment in a competitor, Plenty, just one month after Bowery closed its latest round. Bowery has raised $31 million from VCs. “It definitely gives you pause,” says this investor.
Down to the wire
Still, even as SoftBank barrels forward — Vision Fund has so far deployed $20 billion, including investments in U.S. chipmaker Nvidia and the co-working juggernaut WeWork — questions over its pacing and strategy remain.
Last week, at a small event hosted by this editor in San Francisco, venture capitalist Megan Quinn of Spark Capital acknowledged that there have been times when a company is talking with firms like hers about a sub-$100 million round, and SoftBank has entered the picture “and is like, here’s $200 million!”
Quinn likened the deals to “baby buyouts,” saying that SoftBank is buying more of its portfolio companies than do traditional investors, sometimes because “SoftBank has been able to convince them that something that looks like a baby buyout is actually the right round for them.”
Other times, she added, “the company [that’s fundraising] has not been able to raise from traditional Series C investors, so they’re looking for something more meaningful — someone with more ownership opportunity.”
Asked about these comments, one source close to the Vision Fund tells us, “If you can show [SoftBank] where it can invest $10 million and make a billion, who is going to say no? But how many opportunities are out there like that? If you think about size of the fund and you think about what’s going to move the needle, then it’s fair to say [its team] has to invest a few hundred million in something. Otherwise, it’s not worthwhile.”
Some say that no matter what SoftBank’s aim — whether to produce venture- or instead private-equity-like returns — pouring so much money into tech over a relatively short period is a strategy that’s difficult to grasp.
Speaking alongside Quinn at the same gathering, investor Jules Maltz of Institutional Venture Partners meanwhile questioned SoftBank’s math, saying it “has some challenges.
“If you think about trying to return $100 billion, you don’t just want to get the money back; you want to make a return on that,” said Maltz. “So you probably want $200 billion, meaning you want to double your money, which is good.”
Still, said Maltz, “after fees and all that, SoftBank’s investors [get less than $100 billion in profit]. When you then think that [SoftBank’s] average ownership [stake] is between 15 and 20 percent, and you think about how much liquidity they have to generate in terms of the companies sold in that fund just for them to hit the relatively modest returns of doubling the fund, you start realizing they need more than $1 trillion of market cap in their companies.”
That’s possible if “they get the next Alibaba, the next Facebook,” said Maltz. “And I think that must be what they are going for . . . but . . . I don’t think they can do it.”
Out of the fire
VC Jeff Bussgang of Flybridge Capital Partners seems to suggest it doesn’t matter — that given how fast the money is being put to work, investors should expect that SoftBank will be raising another$100 billion fund soon enough.
“It’s incredibly inflationary when that much capital flows into the system,” says Bussbang, who says SoftBank routinely pays twice the prices that even deep-pocketed mutual fund investors are willing to pay because it can.
Thanks to that 20 percent IRR that it’s targeting, “[Vision Fund] can pay what they’re paying. They’ve promised their LPs a lower return threshold, so they can underwrite to those lower returns, while projecting much bigger outcomes.”
What “people don’t appreciate,” adds Bussbang, is that this is just the beginning. “Son is providing a way for sovereign wealth funds to invest $2 billion at a time into an asset class they can’t otherwise access,” he says. “If they want to invest through a Spark or an IVP, the most [of their capital that they can invest through the funds] is, say, $50 million. Vision Fund meanwhile [accommodates] a flow of capital that hasn’t had a way of accessing this highly fragmented market.”
Unsurprisingly, those close to SoftBank’s thinking say that the fund is simply misunderstood.
Asked, for example, about the giant checks that Vision Fund is writing, and whether these could limit the upside for its portfolio companies’ employees, limit the companies’ exit options or promote the kind of reckless spending that has killed many a promising company, these loyalists argue that SoftBank is anything but careless in its check writing, noting that SoftBank typically demands board seats and certain rights that protect its investments.
“I don’t think it’s overfunding companies,” says one source. “The team is definitely holding companies’ feet to the fire to ensure they allocate their resources the right way.”
It’s also not funding as many companies as the press would have people believe, says this same source, saying specifically of the self-driving car startup Zoox that SoftBank will “not fund it in this lifetime.” (A report last month spawned many more reports that the company was in discussions with SoftBank for a round that could value the company at upwards of $4 billion.)
Says this person, “Zoox has no data, so why would SoftBank fund it?” (Zoox hasn’t yet responded to a request for comment.)
Watch it go, bling!
Despite the massive checks it’s writing, SoftBank has already appeared somewhat fallible at times.
Last week, for example, just one month after leading a $1.1 billion investment in a biopharma holding company called Roivant, a widely awaited drug candidate that would have been a blockbuster for one of Roivant’s subsidiaries, Axovant, was deemed ineffective.
Axovant’s formerly high-flying shares tanked on the news; Roivant remains the company’s biggest shareholder.
It was a crushing development for Roivant. Yet in what could serve as a reminder that it’s too early to judge SoftBank’s strategy, Roivant received more promising news yesterday, when a Phase 3 trial in which another of its subsidiaries is involved produced positive results.
Says one source close to SoftBank, “If mistakes have been made so far, no one will know for another 18 months.”
This person points to Nvidia, whose value has risen since SoftBank invested $4 billion into the publicly traded company in May, when the company’s shares were trading at $137 apiece. (They now trade at $180.) “Whatever is visible,” he added, “has made money.”
Tesla electric semi truck possibly spied in new photo
Is this the new semi truck Tesla is set to unveil at the end of this month? The truck, posted to Reddit (then deleted, then re-posted) bears more than a passing resemblance to the image shared in a teaser released by the automaker itself (via The Verge). The sleek angled front also looks like something you’d expect to be electrically powered, if that makes any sense.
You can compare it yourself to the mostly shadowy teaser image shared by Tesla itself, but the main unifying factor here might be those flared fenders and angled headlights, which you can pretty clearly make out in the Reddit spy shot, too. The photo above was taken “somewhere in California,” where apparently Tesla has been known to test vehicles in the past.
Tesla’s officially revealing its semi truck on October 26, at an event teased by Elon Musk himself. The truck is rumored to have a range of between 200 and 300 miles on a single charge, which would be very impressive for an all-electric heavy-duty transport vehicle, though not something that’s suitable for long-haul trips. The big question will be how it charges – and how fast.
Tesco profit overstatement prompted ‘tears and resignations’
Court hears about ‘stress and worry’ among employees
A number of senior Tesco staff resigned and “people were in tears” because they felt so “compromised” about being asked to “misreport” profits, a court has heard.
Carl Rogberg, 50, Christopher Bush, 51, and John Scouler, 49, are on trial over a £250m accounting scandal in which it emerged that Tesco had overinflated its profits and had to issue a correction to the stock market in September 2014.
The three men all deny charges of fraud and false accounting.
The trial has heard of increasingly frantic attempts to deal with the growing hole in Tesco’s accounts during the first half of 2014 caused by the practice of illegimately “pulling forward” unearned future income.
In August 2014, Amit Soni, a senior accountant at Tesco, was so concerned about the practice that he told Mr Bush, who was managing director of Tesco UK, that the hole in the accounts could reach £600m by the end of the second half, the trial heard.
A number of Tesco staff resigned by August 2014 because they “felt so compromised” — including one senior manager, Richard Parsons, who complained “it has broken me”.
Mr Parsons, a project manager who had worked at Tesco since 2003, became concerned in 2014 about the “goings on” at the retailer, which he did not consider good accounting practice and turned down a promotion to become a commercial director, Sasha Wass, QC, prosecuting claimed to the court.
In August 2014, Mr Parsons sent one email to his line manager speaking of his “stress and worry” and saying he was not “able to talk to anyone outside of Tesco about this, including my wife”.
At his company exit interview, Mr Parsons suggested that “others were too scared to speak out for fear of losing their jobs and paying their mortgage”.
The trial heard an accountant Aysen Nadiri quit without a job to go to because she became “felt compromised as a finance professional” and “felt she no longer wanted to be part of the company”.
Mr Parsons suggested that “people were in tears” and concerned about losing their CIMA accounting qualifications, the court heard.
Mr Soni had already raised his concerns with the three defendants “about the very serious problem of the hole in Tesco’s accounts” but by September 2014 Mr Soni began collating a report on the exact size of the hole in the accounts.
In an email to colleagues in September 2014, Mr Soni confided: “The whistle is about to blow.”
Mr Soni gave his work to the Tesco board in late September, which triggered an investigation.
In September 2014 Tesco had to put out a corrected statement to the stock market detailing the £250m hole in its accounts.
The three men on trial all held senior positions within Tesco. Mr Bush was managing director of Tesco UK, Mr Rogberg was UK finance director and Mr Scouler was UK food commercial director.
They deny wrongdoing. The trial continues.
WHY GOOGLE NEEDS GADGETS
YOU'D THINK HAVING dominated search and email, created Chrome and YouTube, plus a self-driving car project, a handful of save-the-world enterprises, and the greatest advertising business in the history of the universe would be enough to keep Google busy. You certainly wouldn't think the folks in Mountain View would suddenly feel the urge to get into the smartphone game, a remarkably mature market where nobody but Samsung and Apple makes any money, and where Google's already ubiquitous thanks to Android.
And yet, tomorrow, Google will reportedly launch the next generation of its smartphone with the Pixel 2 and the Pixel 2 XL. At the same time, the company will reportedly introduce a new Chrome OS-based laptop called the Pixelbook, a small smart speaker called the Google Home Mini, and new hardware for the Daydream VR platform. The announcements come on the heels of Google's $1.1 billion acqui-hire of 2,000 HTC engineers, who will help Google make more hardware, more quickly. Right or wrong, smart or stupid, Google's a hardware company now.
Of course, Google's made hardware for a long time. The Nexus team built phones; the Pixel team worked on Chromebooks, tablets, and then also phones. The Ara team, within Google's ATAP division, built its own sort of phones. Another team worked on Chromecast, another on Google Wi-Fi, another on the Daydream View. Remember the Nexus Q set-top-box-doorstep thing? That was Google. All these products had the same goal: to show developers and users how good Google's software could be, running on the right hardware. But they were small-time, limited-run products that rarely led to market-wide innovation. In 2016, something finally clicked, and Google took its fate more firmly into its own hands.
As they say, hardware is hard. It's a ruthless and low-margin business, but it's also an important one. Building gadgets in-house gives Google an opportunity to assert itself beyond what any of its partners can offer. More importantly, it gives Google a chance to control its destiny in an increasingly uncertain time. "As new technologies spin out of mobile, Google wants to make sure its own high-end hardware highlights that—whether it’s Assistant or Daydream or Tango, even the Internet of Things," says Avi Greengart, a devices and platforms analyst for GlobalData. "Google needs Samsung, it likes Samsung, but as the platform driver it doesn’t want to be entirely dependent on Samsung."
Depending on Samsung is a dangerous game. Galaxy products are the most popular Android phones by far, and the prime iPhone competition. But every year, you can feel Samsung leaning a little further away from Google. It built the Bixby assistant, which competes directly with Google Assistant, and gave Bixby prime placement on its phones. Samsung builds its own browser, email client, and messaging app, which seem utterly redundant unless Samsung's trying to wean its reliance on Google products. Samsung mostly eschews Daydream in favor of Gear VR, and has a home-grown smart-home platform competing directly with Nest, Android Things, and all the other Google connected-home products. Over the last few years, Samsung's been hammering away at Tizen, its own operating system, which already runs on the company's wearables. Save for the Play Store's un-replicable app selection, Samsung barely needs Android at all.
Soon enough, Android's power might wane for everyone else as well. New platforms like Amazon's Alexa pose something of an existential threat to Google. Alexa can give you directions through Google Maps, or another mapping app. You can listen to Google Play Music, or Spotify. If you buy an Echo, you might someday never use Google, and you might not even notice.
Google could go to the HTCs and LGs of the world, build phones and speakers with them, and try to grow competitors in the wild. Except that doesn't work—Google's been trying for years with the Nexus program to inspire and convince others to build better, cleaner, more powerful Android phones, but everybody just keeps buying Galaxies.
So that's the danger in not making hardware. There's also opportunity in doing so. Just look at Apple, which spent the last decade taking the idea of vertical integration to its extreme. It doesn't just design and assemble its own phones, it's also responsible for many of the most important parts inside. By any measure, the latest crop of iPhones are the most powerful smartphones on the market, thanks to Apple's own A11 Bionic processor. As other companies converge on identical performance metrics, commoditizing a "Good Phone," Apple's jumped to a whole different level. That's why its augmented-reality tech works so well, why its cameras are consistently the best on the market, and even enables minor-but-crucial things like rock-solid Bluetooth connection.
As the next phase of tech comes into the market, that kind of power and control become even more important. "Tiny EarPods, Smart Watches, Augmented Reality, Adaptive Acoustics require wrapping your arms around all parts of the problem," the analyst Horace Dediu wrote recently. "The integration and control it demands are in contrast to the modular approach of assembling off-the-shelf components into a good-enough configuration." Think about the next version of Google Glass (or whatever non-flashback-inducing new name Google gives its face-puter). Making that work will require highly specific work in batteries, processors, screens, cameras, and more. The off-the-shelf version looks like, well, Glass. The custom version, built in-house and in tandem with the software, firmware, and wireless teams at Google, could be much better.
Right now, the company's making a smartphone for the same reason Andy Rubin's Essential started with a smartphone, the same reason anyone talks themselves into entering this preposterously huge market: It's the most important device anyone owns. Google seems to believe it's big and popular enough that it can wedge its foot in the door of the market, and at least take a piece of the high-end smartphone universe. But that's actually the easy part—the last Pixel was great, and HTC made it. The real reason to take everything internal is to be able to control and improve what comes after.
Saying "we're a hardware company!" and actually becoming a successful one are two radically different enterprises, though. That HTC group Google just bought? They've made great phones for a decade, but thanks to some bad carrier-exclusivity decisions and a truly terrible marketing plan, its products stayed under the radar. (Meanwhile, Samsung's marketing department bought the radar and trained it to only recognize Samsung products.) The Pixel, which is a great phone, didn't exactly take the world by storm either. Google hasn't reported exact sales, but one estimate showed a million Pixels in the market; another said between three and five million. That's due in part to Google's total inability to keep the phone in stock—another downside of not managing your own hardware—but doesn't inspire much confidence. Either way, Greengart says, "it certainly hasn’t been a dominant player in the high-end of the market. It doesn’t play at all in one of the biggest markets, China. I can’t see Huawei or Xiaomi or Oppo being worried." Apple and Samsung haven't lost sleep either.
But Google has to try. As we've learned time and again, the best products come from companies that make both hardware and software, each working to optimize and improve the other. Samsung knows it; Apple knows it. As we enter the next phase of tech, where smartphones give way to smart watches and smart speakers and smart lightbulbs and smart cars, there's no room for too-big parts or inefficient software. The winners will be the companies who figure out how to do everything right: hardware, software, marketing, everything. Apple nailed it with phones, and became the richest company in the world. There's another chance coming.
UBER KNEW ITS SELF-DRIVING GURU HAD TAKEN GOOGLE'S TRADE SECRETS, REPORT SAYS
Now a hotly contested due diligence report, commissioned by Uber, makes it clear that the ride-hailing company knew Levandowski had ill-gotten Google files before it bought his startup and put him in charge of its own self-driving efforts. Question one seems to have its answer, and question two just got a lot more interesting.
The firm Stroz Friedberg prepared the report, which Uber used to prep for its 2016 acquisition of Otto, Levandowski’s company focused on self–driving truck technology. Waymo's attorneys filed the report as an exhibit in the case on Monday night, making it public.
Since Waymo filed its suit in February, Uber's battle stance has been: Whatever Levandowski did, we had no part in it. Any stolen files never made it onto our servers or into our cars. No one here used that information to inform how we're developing our own technology.
The due diligence report indicates that Uber’s investigators knew Levandowski had possession of thousands of files related to the Google self–driving car project at least two months after he left the company. The report finds the engineer had access to Google self-driving project design files, source code, laser details, emails, presentations, software, and photos of Google tech and computer screens on his personal laptop, in a Dropbox account that he had used while at the company, and on a set of five disks. But Levandowski had deleted or destroyed many of the files by the time he met with Uber’s investigators, even emptying his computer’s trash while inside the law firm’s offices.
Whether Levandowski conspired with Uber officials to take those documents is still an open question. The report reveals evidence of conversations with Uber executives about working together—a full six months before he left Google in January 2016. At one point, Levandowski asked then-Uber executive Brian McClendon what his company would be willing to pay for the entire Google self-driving staff. (Levandowski said he was trying to pin down a market value for the team.)
Levandowski told Uber’s investigators that he and Kalanick exchanged more than 200 text messages during this period.
Through Waymo, Google is suing Uber for stealing a raft of self–driving car trade secrets and patents, including information on lidar, a sensor that will help autonomous vehicles “see” the world around them. The search giant says that Levandowski stole thousands of documents when he left Google as part of a plan to bring those patents and trade secrets to Uber and use them to jump-start Uber's own self–driving car project. The suit demands Uber pay at least $1.9 billion in damages.
Uber had used every legal maneuver and appeal it could muster to keep the 34-page report out of Waymo’s hands. But on Monday, an Uber spokesperson said the embattled company was “pleased” the due diligence report had been made public. It “helps explain why—even after 60 hours of inspection of our facilities, source code, documents, and computer—no Google material has been found at Uber,” the spokesperson said in a statement.
If you’re still feeling a bit mystified about how and why Levandowski left Google and finally made his way to Uber, know that you are not alone. Levandowski has refused to answer questions, asserting his Fifth Amendment right against self-incrimination. Federal District Judge William Alsup has taken the rare step of referring the case to federal prosecutors to determine whether criminal charges against Levandowski, Kalanick, Uber, and others are warranted. Uber fired Levandowski in May, saying he wasn't cooperating with its legal efforts.
Uber's report makes it clear that Levandowski aggressively recruited for his startup while working at Google. He held one-on-one meetings with more than 20 Google employees at the company’s offices, at recruits' homes, and in coffee shops. At the end of 2015 and in early 2016, he held four recruiting meetings at barbecues at his house and on a ski trip to Lake Tahoe. (A number of Google employees were invited to these soirees through their work email addresses.) Levandowski told Uber’s investigators that his startup made employment offers to at least 15 Google employees before and after his departure from the company. By mid-March 2016, Otto had 30 employees, 16 of whom were former Google workers.
More curious is the business with Levandowski’s five disks, which contained, the report says, proprietary information. The engineer told Uber investigators that he discovered the disks inside a closet sometime just after he suddenly resigned from Google. By Levandowski’s retelling, he immediately informed his attorney about the disks and alerted Uber’s top brass. An Uber executive told Levandowski to preserve the disks for record-keeping purposes, but by the time he met with Stroz investigators, Levandowski said he had taken them to an Oakland shredding facility to be destroyed. The subsequent investigation did not turn up any hard evidence that the shredding facility destroyed the five disks but suggested that if it did, it only happened three days after Levandowski met with investigators.
Waymo’s legal team is touting the report’s revelations as a victory. “Knowing all this, Uber paid $680 million for Mr. Levandowski’s company, protected him from legal action, and installed him as the head of their self-driving vehicle program,” a Waymo spokesperson said in a statement. “This report raises significant questions and justifies careful review." The team has used the report to ask Judge Alsup for a delayed trial, which would give the company more time to review the document and others recently released by Uber.
But the company still hasn’t firmly linked Uber to the stolen files—that’s that pesky question two. Waymo will need to prove that Levandowski’s information made it into Uber self-driving tech, or that the ride-hailing giant was negligent enough to make that leakage possible.
“So far you don’t have any smoking gun,” Judge Alsup told the company's lawyers in May. That was thousands and thousands of documents ago. Waymo’s team has promised there are bombshells to be revealed, but for now, the air in the courtroom looks pretty clear.
Telcos angry as European Parliament waters down investment plans
Brussels politicians’ vote undermines case for fibre and 5G deployments
The European telecoms industry has reacted with fury after the European Parliament watered down proposals designed to encourage investment in 5G and full fibre networks and pushed to introduce regulation of international calling rates on the continent.
Its vote undermines proposals put in place by the European Commission last year to encourage companies to invest in faster telecoms networks with confidence, in exchange for deregulatory measures. In mobile, the European Commission has estimated it could cost €500bn to push the continent to the forefront in the race for 5G networks, but that a funding shortfall of €155bn from the industry imperils that goal.
Yet the European Parliament has amended the proposals to include language aimed at tackling “oligopolies”, where a small number of large telecoms companies have too much power. It also failed to guarantee that companies investing jointly in fibre infrastructure would benefit from deregulatory measures. In addition, it proposed handing more power to national regulators, despite the push for a digital single market.
Barclays analysts said that its vote had replaced “carrots” with “sticks”, compared with the original positive proposals by the European Commission. “Worryingly, a number of the pro-investment measures appear to be watered down, which we see as a potential negative and barrier to investment,” the bank said in a research note.
Lise Fuhr, director-general of ETNO, the trade body that represents Europe’s largest telecoms companies, said: “Today’s strategic challenge is network investment, but this parliamentary vote misses the point and risks to slow down broadband deployment.”
Vodafone, which has invested in fibre in Spain and Portugal and is in talks with Openreach in the UK over co-investment plans, said the revised proposals increased the “regulatory burden” on companies. “Those amendments if adopted would erode investor confidence in European telecommunications, reducing further the prospect that EU businesses and households will benefit from the competitive gigabit-speed networks that will be critical to all aspects of daily life and work in the near future,” a spokesman said.
The bill will be negotiated with the member states into early 2018 and could change further during the talks. There is the prospect of a further watering down of measures, including the European Commission’s plea to introduce spectrum harmonisation across the continent, something opposed by member states. Andrus Ansip, the Commissioner for the Digital Single Market, said last week at the FT ETNO conference that Europe faced an “emergency situation” due to a lack of progress.
The vote by the European Parliament also opened up a new frontier in retail price regulation for the telecoms sector as it proposed capping the cost of “abusive” international calls between countries within the European Union. The move comes after the lengthy push to abolish roaming fees.
Telecoms companies believe it may be hard to resist the political pressure to introduce caps on international calls now it has been worked into legislation.
Mr Ansip argued that the European Commission’s plan to introduce such legislation was rejected by the parliament in 2013, which he said was a “wise” move given consumers increasingly use apps including WhatsApp, FaceTime and Skype for video calls and messages. “If members of the European Parliament feel we have to deal with this rapidly declining market then we have to, but I don’t think it is more reasonable to deal with it today than in 2013,” he said.
Privacy is under threat from the facial recognition revolution
Without protection, the rights of citizens and consumers will wither
We have grown used to cameras watching us. Once they recorded us; increasingly they recognise us, too. It is easy to think of facial recognition as just a technological convenience — a new way to unlock a smartphone, say. That is a mistake. In essence, facial recognition is biometric identification at a distance, with no need for consent.
Unlike fingerprints, retinal scans or blood samples, it is easily performed without the subject’s knowledge. It will affect how we travel, live, shop, and much else. It will force changes in the way privacy is defined and protected. If those who care about individual rights do not start thinking about the implications now, those changes will be forced upon us rather than chosen.
Facial recognition is already in use around the world. The Chinese equivalent of Amazon, Alibaba, allows people to “pay with a smile” using facial recognition in stores, for example. The potential for good is obvious. Think of the hours that could be saved if facial recognition were to become the default identification tool at airports.
These benefits will have to balanced against the loss of anonymity. In Russia, an app called FindFace identifies individuals in photos, linking them to profiles on a social network called VKontakte. A similar service, if linked to Facebook and other networks, could put names to billions of faces. In the city of Shenzhen jaywalkers are identified using CCTV, and their faces and addresses posted on a large screen to shame them into better behaviour.
The technology will not be limited to connecting a face with information already present on the internet. A facial recognition model developed at Stanford, when presented with paired photos of individuals who self-identify as gay or straight, could tell which was which with 81 per cent accuracy in men and 74 per cent in women. Humans given the same task were much less accurate. Yes, the sample was limited and the study needs to be replicated with a more refined methodology. The results cannot be dismissed, though. Nor can the frightening implications. Consider an algorithm identifying sexual minorities deployed in an intolerant, authoritarian state. The technology may misclassify many, but tyrants lose little sleep over false positives.
Facial recognition also may appear in a store near you, in the form of “dynamic pricing”. Supermarkets are replacing paper price tags with digital screens to avoid manually relabelling for price changes. Imagine if the prices changed based on the identity of the shopper, picked out by a camera at the store entrance. Does this sound too farfetched? Travel sites allegedly charge more when they detect the search patterns of a determined buyer. Amazon already changes prices quickly on the basis of demand. Economists call this price discrimination. With a set price, consumers who could have paid more capture surplus value. In a world of tailored pricing, this discount vanishes, increasing returns for sellers at the expense of buyers.
Dynamic pricing might be possible even without positive identification of the target. A University of Toronto study suggests that facial cues can provide useful hints about socio-economic status. A facial identification algorithm that detected class would be a hot commercial property.
Facial recognition technology will develop apace, whatever lawmakers and rights advocates do. When citizens have the right to anonymity is no longer a theoretical question. If the technology becomes ubiquitous before decisions are taken about legal limits, it may be the facts on the ground, rather than liberal principles, that determine the answers.