Fighting Instagram’s $1.3 Billion Problem—Fake Followers

https://www.wired.com/story/instagram-fake-followers

On Instagram, manufacturing fake followers is a ubiquitous tactic, one that’s churned out at least 95 million near-perfect human forgeries for you to brush past in the digital hallways, unaware. You can buy them in droves from dozens of services online, or even from a coin-operated vending machine created by artist Dries Depoorter. He calls his fame dispenser “Quick Fix“: a wall-mounted box with an Arduino and a keyboard, where visitors can type in their social media handles and select what faux-honorific they’d like to receive—likes or follows, starting at just one euro, delivered instantly. In Helsinki, where the machine debuted, “Quick Fix” was a revelation. “Even the young people didn’t know this was possible, but they wanted it,” Depoorter says. “In a city, I think it would fit really well next to a soda machine.”

The fake followers would fit in anywhere: padding out an obscure microinfluencer’s ranks, creating the illusion of customers for would-be T-shirt seller, skulking among legions of real fans on celebrity accounts like Ellen DeGeneres’. A report by cybersecurity firm Cheq projects that these fake fans will cost brands $1.3 billion in 2019 alone. In the influencer economy, you’re paid in part for the size of your audience, which means that, if an influencer’s account is swarming with fake followers, brands are paying extra to reach people who don’t exist.

Agencies and would-be sponsors know they’re being scammed, and are growing more vigilant. “In the last few years, we’ve started holding influencers accountable in their contracts,” says Gabrielle Vogt, senior manager of digital talent at influencer marketing agency HelloSociety. “They have to agree that they haven’t participated in comment pods, botting, or purchasing fake followers.” Let’s break down these forms of fakery. “Comment pods” (or “engagement pods”) are groups of influencers who agree to like and comment on each other’s posts to artificially drive up engagement and improve their algorithmic performance. “Botting” is using automated fake accounts—bots—to bolster an audience, à la the Kremlin-linked Internet Research Agency. In their contemporary form, “fake followers” are often real people, just not always the person they say they are, and, of course, not someone who is an actual fan (and hence, potential customer).

Not being rubes, influencer economy participants know that signing a contract isn’t enough to certify continuous truthfulness. That’s why HelloSociety and most other influencer marketing businesses rely on online tools to sniff out counterfeit or fraudulent accounts—and influencers who profit from them. All of these tools (and there are many, with names like Social Audit Pro, IG Audit, Hypr, HypeAuditor, and Famoid) work a little differently, but they’re all looking for deviations from the platform’s norm. An account with no profile picture, say, or one that follows 10,000 people and never posts. Also suspicious: lots of digits in the username; or a person who seems to live in Turkey or Indonesia, but follows only California-based influencers.

Online tricks evolve like Pokémon, though, and fake followers are getting much harder to identify. The best way to spot fakes now seems to be taking note of the rate at which an influencer acquires followers. “A year ago, I would see influencers buy one to ten thousand followers at a time, and it would show up on a graph as a giant jump out of nowhere,” says Andrew Hogue, creator of IG Audit, an analytical tool he describes as “Carfax for influencers.” (Vogt notes that those jumps can happen naturally, like after a Today show appearance, so HelloSociety always checks the graph against an influencer’s post history.) Lately, sellers have been offering “drip followers”: If an influencer orders 30,000 followers, for example, they arrive only 50 at a time, every day, for 600 days. According to Hogue, detecting drip followers requires a closer inspection of those follower account graphs. “Normal users will have a smooth logarithmic slope,” he says. “If you’ll see a jagged, stair-like pattern out of nowhere, that’s fraud.”

The need for vetting is so strong that the demand caught the makers of several tools by surprise. “I’ve never paid a cent for marketing,” says Robert Harris, creator of Social Audit Pro, a service that analyses the quality and activity of online followings. “It went viral on its own.” When Hogue debuted IG Audit, he saw it go viral in Poland overnight before catching on in the United States, where its virality continues. Last month, the Institute of Contemporary Music Performance ran 100 of the most successful Instagram accounts through IG Audit, reporting that up to half of all the people following celebrities like Ellen DeGeneres, Korean boy band BTS, Kourtney Kardashian, soccer player Neymar, Ariana Grande, and Taylor Swift weren’t real followers at all. The list was written up as news in at least 10 publications.

via Wired Top Stories https://ift.tt/2uc60ci

September 10, 2019 at 07:03AM

The new battle in Hong Kong isn’t on the streets: it’s in the apps

https://www.technologyreview.com/s/614286/the-new-battle-in-hong-kong-isnt-on-the-streets-its-in-the-apps/

Alice had been marching for hours by the time she got home on the evening of July 21. It was a Sunday, and the sweltering heat of a Hong Kong summer had left her exhausted and jittery. Still, she was transfixed by her phone, which had been buzzing the entire subway ride.

It was pinging with news alerts about a group of protesters who had split from the peaceful procession and were facing off against riot police. On arriving at her apartment, she opened her laptop and fired up a website that displayed a dizzying nine separate live shots from different news outlets. One stream in particular caught her eye, showing chaos at a nearby railway station. She could see men armed with clubs who were terrorizing train passengers, bursting into subway cars and bludgeoning bystanders who were kneeling on the floor begging for mercy.

The live stream was gripping and horrifying: Alice felt as if she were in the middle of the station. She flinched and screamed when an attacker in a pink shirt hit a reporter who was filming from her cell phone. Despite being beaten to the floor by the blows, the journalist kept on filming and narrating the scene as she got back on her feet. When the police arrived, Alice saw passengers screaming at them for turning up only when the violence was over. The crowd’s anger swelled, and the officers eventually retreated.

“I know I wasn’t there,” Alice, who asked to be identified only by the English name she uses because of fears for her safety, told me recently. “But the experience was deeply visceral. I could feel the mood and the tension in the station, of how angry everyone was at the police in real time. There were moments where it felt like the police would also start beating people. And the reporter narrating, shouting questions … she sounded like she could be me.”

Information, streamed

Hong Kong is famous for its souk-like electronics malls, and it’s blanketed with high-speed internet. So when protests broke out in June over plans to implement a controversial extradition law—which would see Hong Kongers accused of crimes turned over to mainland China’s notoriously opaque justice system—it was natural that many people turned to online services for more information and guidance.

Some of these methods have already been heavily documented. Everything from supplies of food and water to press conferences are put together in the chat app Telegram, for example. Meanwhile, LIHKG, a Reddit-like forum that is limited to local ISPs, provides a sandbox of ideas where a network of anonymous citizens can exchange memes, protest schedules, and tactics. Online polls often dictate the location of the next traffic-disrupting flash mob.

And, as in many protests in the past, a small army of journalists and activists have been live-streaming everything from major marches to minor spats with police. The raw videos tap into local media habits—many people leave live streams playing in the background while they cook dinner or hang out with friends—and help create a sense of solidarity and belonging, even among those who are not on the streets themselves.

Supply chain: Thanks to messages on Telegram and information sent via AirDrop, protesters are able to get supplies to the front lines through chaotic scenes.

An Rong Xu

(Some even watch live streams while they are attending protests: At one recent demonstration, an elderly man watched a video feed showing a group about 50 yards away. As they sang a rhyme taunting police, it echoed a moment later on his cell phone.)

Gwyneth Ho, the reporter for Stand News who was attacked for holding the camera that Alice was watching that night, says it creates a very direct connection for many viewers.

“We disregard quality and framing, but we’re in the middle of the protesters and even the police, and people get really immersed in the scene,” she says. “The audience doesn’t want well-packaged shots—they want to feel what it’s like to be on the ground, in the most dangerous situation.”

She adds, “A lot of people have told me it was like a VR experience of getting beaten.”

Heavy hand: More than a million people attended a rally condemning police brutality in August.

An Rong Xu

Real time

Video footage has been important to protest movements many times before, of course, and social media and online messaging have been influential elsewhere as well—including the Arab Spring protests that spread around the Middle East and North Africa in the early 2010s. But Hong Kong has developed some of its own techniques, too.

Several teams of volunteers have started constructing and sharing live maps to help those on the ground during demonstrations. They’re the brainchild of a man who goes by Orca, an educator in his 40s. He was spurred into action after seeing “massive panic and lots of anxiety” during one protest that gave rise to clashes in a high-end shopping mall (“No one knew where the police were or how they could get to an escape route,” he told me. “So our team began planning to map out the next big rally the following week.”)

Now Orca and his team publish dozens of maps during large demonstrations, updating positions with colors to show the location of police, “thugs,” and protesters, plus icons to signify first aid, rest, and supply stations. All of this is put together by on-the-ground volunteers who draw the information out on a blank map on their iPads, and send it to an “integrator” who compares the data with news from live streams and television stations before putting it all together and sending it out over Telegram or Apple’s AirDrop file transfer service. During one rally, an estimated 600,000 people downloaded maps put out by Orca’s team, just one of three mapping services created during the protests.

Alice—who has used Orca’s maps on several occasions—never thought her involvement would extend beyond the occasional march. But something changed after she watched the subway rampage.

Live action: Demonstrators are using real-time maps, with information sent in by volunteers, to track activity inside the huge rallies.

An Rong Xu

In the days that followed, she spent her daily commute AirDropping protest art and information about the attack to anyone with an open connection. A week later she marched in her first unauthorized assembly, in the suburb where the train attacks unfolded.

She began donating the little money she could spare to an online fund-raiser helping pay arrested protesters’ legal fees, and after being tear-gassed by police herself, she donated boxes of gas mask filters. She would leave spare change on top of subway ticket machines, allowing protesters to buy single-use tickets to avoid being tracked.

Then, three weeks after she watched the train station attack, Alice decided her contributions needed to become more direct. During one of the most violent weekends so far, she joined the crowd, carrying a rucksack filled with supplies: bandages, water, snacks, and filters for gas masks. When she saw a call on Telegram, she rushed forward toward police lines for the first time, opened her bag to those in need, and quickly retreated, checking Orca’s maps to avoid running into police.

New battlegrounds

On August 24, Alice looked over a crowd of demonstrators gathered around one of the 50 smart lampposts that had been installed around the city since June. Each one—and another 350 are on order—is stuffed with cameras and surveillance equipment. Posts on Telegram had told her about the action that was set to unfold, and she watched as other activists took power tools to its base.

Alice could no longer call herself a passive participant. She was dressed in what has become the uniform among demonstrators: black from head to toe, her face obscured by a black surgical mask and a black baseball cap.

The towering metal pole fell with a loud clunk, and the crowd erupted in cheers. Protesters immediately descended on the lamppost and began pulling out components, photographing manufacturer and component details and uploading the information. Demosisto, a pro-democracy political party, published a quick analysis of each component.

Lokman Tsui, an assistant professor at the Chinese University of Hong Kong and the former head of free expression in Asia at Google, says that while technology has been crucial in building support for the movement, many have shifted the focus of their efforts to outsmart the government. “In the past few months people have educated themselves incredibly quickly on end-to-end encryption, only buying single-use transit cards, and the dangers of widespread surveillance,” he says.

Police have arrested over 1,100 people in the last three months and have become increasingly aggressive in their tactics. Protesters are worried, and their behavior—from attacking CCTV cameras by spray-painting them or smashing them with metal poles to finding ways to avoid communicating with each other over unencrypted services—reflects their reaction to this situation.

On demand: Protestors have been planning their moves online—but are concerned about being surveilled by the authorities.

An Rong Xu

Tsui coauthored a paper this year showing how many types of data Hong Kong’s telecommunications companies do not consider to be personal and protected, including a user’s geolocation and IP addresses, as well as the information on websites visited. This interpretation, which was made privately by the companies themselves and has not been challenged in court, means that police do not need a warrant to request, say, a list of subscribers who were in a certain place at a certain time. Information collected by Hong Kong authorities could also be handed over to China, Tsui added, since there is no formal agreement defining what can and cannot be shared.

Protesters have become so wary of sharing any identifying information that no one directly involved in the protests agreed to be identified by name. Orca would only conduct an interview over Telegram; Alice asked she be referred to by her English name, which is not on her official ID card.

Alice does not even know the real names of several friends she’s made at the protests. When they message on Telegram, they use their aliases—all English pseudonyms. Even though they are anonymous, anyone who is arrested is cut out of the group for fear that police could compromise their phones.

Five demands

Carrie Lam, Hong Kong’s chief executive, recently announced she planned to formally withdraw the extradition bill, meeting one of the protesters’ five key demands. (Their other requests: Lam’s departure, democratic elections, an independent investigation into police actions, and amnesty for those already arrested.)

But the move seems unlikely to placate the public.

Protests continue, with privacy issues and the police—who have faced many allegations about excessive use of force—now the focus of anger. The mood was felt almost immediately after Lam’s announcement, not in people taking to the streets but in online posts echoing a popular protest chant: “Five demands, not one less.”

With no end in sight, Lam has considered invoking emergency powers, according to local media. One of her first targets would likely be the apps that protesters use to organize. The mere suggestion was so divisive that members of Lam’s cabinet warned her against the move, and the Hong Kong Internet Service Providers Association declared that “any such restrictions, however slight originally, would start the end of the open Internet of Hong Kong.”

Nowhere to run: An unnamed protestor scrambles to get out of Yuen Long, the suburb where Alice witnessed several confrontations.

An Rong Xu

The Chinese government’s concern is that the internet is also the most likely way the Hong Kong protests could spread to the rest of the country. This scenario terrifies Chinese leaders: strict censorship rules have been issued, and border guards regularly check the phones of people traveling from Hong Kong for any sign of protest photos or videos. After Lam’s announcement that she would withdraw the extradition bill, posts on Chinese social media wondered why those elsewhere in China face jail time for even a hint of dissent. The messages quickly disappeared, and search results were replaced with a message warning that the query “does not comply with relevant regulations.”

But amid attempts by the Chinese government to deter protesters by releasing viral clips on Twitter threatening a military crackdown, there is little sign Hong Kongers are cowed. Alice feels that their collective efforts are leveling the playing field between the government and demonstrators.

“When the government lies to the people every week, every day, we cannot trust their promises or that they will follow the law,” she says. “This moment is our last chance to fight for Hong Kong, or the next generation won’t even know what privacy is.

“The government uses an old playbook, but we have created whole new ways of resisting. And if we didn’t stand up and [we] let Hong Kong become just another Chinese city, all that creativity would be snuffed out.”

via Technology Review Feed – Tech Review Top Stories https://ift.tt/1XdUwhl

September 10, 2019 at 04:07AM

Amazon employees are going to strike over the firm’s climate policies

https://www.technologyreview.com/f/614290/amazon-employees-are-going-to-strike-over-the-firms-climate-policies/

It’s part of a global general strike due to take place on September 20, led by activist Greta Thunberg.

The news: Over 900 employees have signed an internal petition to walk out over their employer’s lack of action on climate change, Wired reports. It would be the first time workers at its Seattle headquarters have gone on strike (however, Wired also reports that many of them are taking paid vacation to do so, which somewhat misses the entire point.)

It’s likely the number planning to strike will grow, as Amazon workers in other offices around the world have expressed an interest in joining. Employees at Microsoft are also planning to walk out. There have been previous strikes at Amazon, but they have mostly been conducted by workers at the company’s warehouses demanding better conditions and pay.

The demands: The group behind the strike, Amazon Employees for Climate Justice, have issued three demands. It wants Amazon to stop giving money to politicians and lobbyists that deny climate change, to achieve zero carbon emissions as a business by 2030, and to stop working with oil and gas firms to accelerate extraction.

Growing activism: There’s been a significant growth in protests by people working for tech giants in the last couple of years, and revelations from Gizmodo in April over how Amazon has been courting the fossil fuel industry were met with anger from employees. Amazon has said it will make 50 percent of shipping net zero carbon by 2030, but that is not fast enough for the protestors. Amazon has also promised to reveal its carbon footprint at some point this year, but has yet to fulfil that pledge. For now, we can only guess.

This story first appeared in our daily newsletter The Download. Sign up here to get your dose of the latest must-read news from the world of emerging tech.

via Technology Review Feed – Tech Review Top Stories https://ift.tt/1XdUwhl

September 10, 2019 at 06:22AM

Europe’s ArianeWorks Aims for Reusable Rockets (with a Very SpaceX Look)

https://www.space.com/arianeworks-reusable-rockets-themis-callisto-video.html

Europe’s rocket-launching industry is gearing up to go reusable.

The European launch provider Arianespace — best known as the manufacturer of the heavy-lift Ariane 5 and the future Ariane 6 — has a plan to make its future rockets more competitive in a tight launch industry. As you might guess from looking at the U.S. company SpaceX’s reusable Falcon 9 and Falcon Heavy rockets, reusability is what Arianespace wants to do as well.

Back in February, ArianeGroup and CNES (the French space agency) signed a memorandum of understanding for a new “acceleration platform” that will work to develop new launchers, including reusable ones. The platform, called ArianeWorks, unites teams under one roof and provides all the ingredients possible for innovation: “a highly flexible environment, open to new players and internationally,” according to a press release from the time.

Video: ArianeWorks’ Reusable Rocket Plans Explained
Related: The Evolution of SpaceX’s Reusable Rockets in Pictures

An ArianeWorks reusable Themis rocket prototype returns to Earth in a video animation of the group's new reusable rocket program.

An ArianeWorks reusable Themis rocket prototype returns to Earth in a video animation of the group’s new reusable rocket program.

(Image credit: ArianeWorks)

The interim results are coming soon: two low-cost demonstrators that will examine how to recover the first stage of a rocket launching to space. An elementary experimental vehicle called Frog will test many of the technologies needed for this kind of work, including “landing algorithms, automated operations and the avionics architecture.” These concepts will then be ported on to another, more robust demonstrator called Themis. 

In May, ArianeWorks announced that a Paris-based prototyping firm called MyCTO would build the first Themis prototype; the partners on Frog include Planète Sciences, Polyvionics and Cachan Technology Institute (Paris Saclay University).

ArianeWorks will use the small Frog reusable rocket demonstrator to test technologies for larger reusable rockets.

ArianeWorks will use the small Frog reusable rocket  demonstrator to test technologies for larger reusable rockets. 

(Image credit: ArianeWorks/MyCTO)

CNES and ArianeSpace are also working together to make an engine called Prometheus, which uses oxygen and methane as its propellant and can be adapted for multiple rocket platforms. Methane and oxygen produce products that are more environmentally friendly than many other rocket fuels. Themis will use the Prometheus engine for its landings, which means that the rocket demonstrator will not only be reusable, but also less harsh on the environment during launch and landing.

This isn’t the first time that Arianespace has looked at reusable vehicles; a reusable sounding rocket called Callisto is also in the works. 

Callisto is meant to test how fuels perform in various engine designs. Some of the groundwork for Callisto will flow into the development for Themis, Arianespace said in a YouTube video released earlier this year. 

Follow Elizabeth Howell on Twitter @howellspace. Follow us on Twitter @Spacedotcom and on Facebook

Have a news tip, correction or comment? Let us know at community@space.com.

via Space.com https://ift.tt/2CqOJ61

September 10, 2019 at 11:10AM

Some Chromebooks mistakenly declared themselves end-of-life last week

https://arstechnica.com/?p=1564377

If your new Chromebook told you it's past EOL, don't panic—as long as you were running Canary or Dev, that is. (If you were running Stable or Beta, you may continue panicking.)
Enlarge /

If your new Chromebook told you it’s past EOL, don’t panic—as long as you were running Canary or Dev, that is. (If you were running Stable or Beta, you may continue panicking.)

A lot of Chromebook and Chromebox users don’t realize this, but all ChromeOS devices have an expiration date. Google’s original policy was for devices to be supported for five years, but the company has recently extended that time to 6.5 years.

When your Chromebook or Chromebox approaches its built-in expiration date, it will warn you that it’s time to go buy a new device entirely. Not long after that, it will refuse to apply any further security or feature updates. In addition to leaving users vulnerable to unpatched security exploits, this means that constantly evolving services such as Gmail will eventually stop working entirely.

Google has been working on a way to warn users six months ahead of time that their device’s EOL date is approaching to allow them to plan with a little less time-sensitive desperation. But users running the Canary and Dev early-preview ChromeOS builds discovered a bug in the new code the hard way. After any reboot, brand-new devices started warning “this is the last automatic software and security update for this Chromebook. To get future updates, upgrade to a newer model.”

Luckily, the code had not yet made it to stable—or even beta—builds, so few users were exposed to it. The bug has already been patched, and users on Canary should already have seen the bogus message disappear after a daily upgrade push this past weekend. Users on the Dev build only get updates weekly, so they might not have the warning fixed until sometime this week.

On the other hand, if you’re not running a Canary or Dev build and you’ve gotten a message about your device being beyond its support date and getting no further upgrades… it’s legit, and you really won’t be getting any more upgrades. This doesn’t have to be the end of the world; if you’re willing to get your hands dirty, you can still install Linux as a fully bootable OS, which will never expire on you, using chrx. (This is how I get Linux on my own small fleet of Chromebooks, which I use for multi-station Wi-Fi testing.)

You can also install an alternate version of ChromiumOS itself, such as the freely available Home Edition of Neverware CloudReady. This gives you an experience extremely similar to full ChromeOS, built on the same technology, which will continue to get software updates despite your device’s “expiration date.”

via Ars Technica https://arstechnica.com

September 9, 2019 at 04:55PM