Trump’s Twitter Following Jumps, Sparking Fears of a Bot War

Twitter has been one of Donald Trump’s key tools ever since he started running for president, so it’s no surprise that his account is closely followed by millions of users, and his every tweet is parsed for intelligence about his intentions.

More recently, however, attention has focused on the fact that Trump’s already large Twitter following appears to have increased substantially, and that many of these new followers seem to be “bots” or automated accounts. This has triggered fears in some circles of a looming “bot war.”

According to a number of tweets, including one that got a lot of traction on the social network, Trump’s following increased by as much as 5 million in a matter of days, and most of the new followers were automated or fake accounts.

A spokesman for Twitter, however, told BuzzFeed this report was not true, and a comparison that the news site did between Trump’s current personal Twitter profile and an archived version of the page shows that it has only increased by about 300,000 in the past few days.

While it may not have grown by several million in just a few days, Trump’s account has added more than 2 million followers this month, according to Mashable. It has added about 7 million since February, the site said, and more than half of them have blank profiles.

A report from a third-party service called Twitter Audit shows that more than half of those following the president’s account are suspected of being fake or automated (although BuzzFeed notes that the site’s methodology for detecting fakes is not foolproof).

These large increases, and what appears to be a huge number of automated followers, have sparked a number of theories, including one from Newsweek that suggested the president or his team might have purchased fake followers, something celebrity accounts occasionally do.

But another theory being promoted by some observers is that the bots are part of a deliberate build-up for a forthcoming “bot war” between the Trump administration and its critics, one that might be part of a Russian attempt to influence public opinion related to the president.

Malcolm Nance, a retired U.S. Navy cryptologist and intelligence expert, said he believes that the increase in bots following Trump could show that “Russian cyber warfare support [is] ramping up for @POTUS” and that this kind of behavior is a “key intelligence indicator.”

Intelligence sources working for a number of agencies including the FBI have suggested that agents working for or with the Russian government may have tried to influence the outcome of the U.S. election. And Facebook recently released a report that indicates organized groups tried to influence public opinion via fake accounts sharing hoaxes or misinformation.

Hillary Clinton Blames the Russians, Facebook and Fake News for her Loss

Looking back on her failed election campaign, former Democratic candidate Hillary Clinton said she takes responsibility for every political decision she made, but “that’s not why I lost,” she told attendees at the Code conference in San Francisco on Wednesday.

In addition to press attention focused on her use of a personal email server—which Clinton called a “nothing-burger” that the New York Times “covered like it was Pearl Harbor”—the former Secretary of State said she was subjected to an unprecedented campaign of fake news and social engineering on Facebook, orchestrated by Russian agents and an army of bots.

Clinton said that while her campaign was using social media to reach out to potential voters and supporters, Republican groups were engaged in the “weaponization of technology” to push a message about her and the risks of electing her president.

“Here’s what the other side was doing,” she told interviewers Walt Mossberg and Kara Swisher of Recode. “Through content farms, through an enormous investment in falsehoods, fake news, call it what you will—lies—the other side was using content that was just flat out false, and delivering it both above and below the radar screen.”

In terms of Facebook, Clinton said that the “vast majority” of news items that appeared on the social network about her were fake. This orchestrated campaign was “connected, as we now know, to a thousand Russian agents [and] connected to the bots, which are just out of control,” she said.

Clinton referred to a recently declassified report from the Director of National Intelligence, which said that a number of intelligence agencies agreed there was Russian involvement in the campaign.

“Read the declassified report that came out in early January,” Clinton said. “Seventeen agencies all in agreement—they concluded with high confidence that the Russians ran an extensive information war campaign against my campaign to influence voters in the election.”

These fake news stories helped convince potential voters not to support her, Clinton said. And the former Secretary of State said she and her campaign were convinced of Russian government involvement in a disinformation campaign and other dirty tricks early on.

“We went and told anyone we could find that the Russians were messing with the election and we were basically shoo’d away,” Clinton said. “We couldn’t get the press to cover it.”

Clinton also referred to Cambridge Analytica, a data-analysis company that specializes in using demographic and psycho-graphic data about online behavior to target political and advertising messages. Some have credited the firm—which is controlled by Robert Mercer, a prominent backer of the Trump campaign—with helping to sway the election.

The former Senator said that it’s important for people in the technology world and the business world to understand the connections between “domestic fake news operations” and the sophisticated attempts by Russian cyber agents to influence user behavior.

“How did they know what messages to deliver?” Clinton asked, referring to the Russians. “Who told them? Who were they colluding with?” She also called on Facebook to do more about fake news on the network. “They’ve got to curate the news more effectively,” she said. “They’ve got to help prevent fake news from creating a new reality.”

Clinton called her use of a personal email server a “nothing burger,” that got turned into “the biggest scandal since Lord knows when.” New York Times covered it “like it was Pearl Harbor.” She said “I take responsibility for every decision I made, but that’s not why I lost.” Talks about the “weaponization of technology” by various groups, as well as the Russians and Cambridge Analytica. “Here’s what the other side was doing — through content farms, through an enormous investment in falsehoods, fake news, call it what you will — lies — other side using content that was just flat out false, delivering it both above and below the radar screen — look at Facebook, vast majority of the news items were fake — connected as we know now to 1,000 Russian agents, connected to the bots, which are just out of control, see reports now about Trump’s account and all the fake accounts following him…

Clinton noted that she had no control of the Russians, which she said “weaponized” technology against her. She cited the deluge of false articles that circulated on Facebook in the months preceding the election that were “connected to the 1000 Russian agents,” and WikiLeaks, which spent the month before the election releasing a daily trove of emails from her campaign Chairman John Podesta.

Google searches for WikiLeaks, she said, were highest in Wisconsin and Pennsylvania, swing states crucial to her victory that she narrowly lost. But, she said, the Russians couldn’t have acted unilaterally; they had to have had support from Americans. “How did they know what messages to deliver?” she asked about the Russians. “Who told them? Who were they colluding with?”

“It’s important for people in tech and business to understand the marriage of the “domestic fake news operations,” the sophisticated Russian cyber units and the Republicans’ more flush data repository, Clinton said.

“Putin wants to bring us down,” Clinton said. “It’s way beyond me. …. I believe that what was happening to me was unprecedented. Over the summer we went and told anyone we could find that the Russians were messing with the election and we were basically shoo’d away. …. We couldn’t get the press to cover it.”

Clinton said platforms like Facebook have got to get better at curating news. But she also said that her supporters put off taking more action on fake news because she was thought to be in the lead. “I don’t know enough about what they could have done in real time,” Clinton said. “I also think I was the victim of the very broad assumption I was going to win. I never believed it, I always thought it would be a close election.”

Bill Simmons Is Moving His Sports Media Empire Again

When sportscaster Bill Simmons left ESPN and subsequently started a new publication called The Ringer, in partnership with the blogging platform Medium, it was seen as a huge loss for Disney-owned ESPN. Now, Simmons is on the move again — this time to Vox Media.

Vox CEO Jim Bankoff told the Wall Street Journal he sees Simmons and The Ringer as a strong brand that will fit well with the company’s portfolio. Vox Media, whose financial backers include NBCUniversal, operates a range of sites including The Verge, Recode and the sports-blog network SB Nation.

Simmons will retain ownership of The Ringer and its various offshoots, which include a successful podcast that he started while at ESPN. Bankoff said Vox and Simmons will share the advertising revenue from those assets, although the exact ratio was not made public.

Before he left in 2015, Simmons was one of the stars of the ESPN universe, thanks to the audience he had built up for Grantland, the standalone sports and culture site he created for the network. His departure was seen by many as a sign that ESPN was losing traction with sports fans, and he made no secret of the fact that the parting was acrimonious.

Sources at ESPN, meanwhile, said at the time that Grantland was mostly a vanity project for Simmons and didn’t really generate enough traffic or advertising revenue to make it worthwhile for the site to continue, in part because of Simmon’s $3-million salary.

After leaving the sports network, Simmons signed a high-profile deal with HBO to produce and star in an interview show called Any Given Wednesday. But despite his passionate following in the sports world, the program failed to generate much positive buzz, and it was eventually shelved just four months after it launched.

When the show launched, HBO became an investor in Simmons’ holding company, which produced the show and his other ventures. It’s unclear what role the broadcast network will have in the current relationship with Vox, or whether Simmons will continue working with HBO as well.

Several months after he left ESPN, Simmons announced that he was starting a new venture called The Ringer, and that it would be hosted by Medium, the blogging platform founded by Twitter co-founder and former CEO Evan Williams. At the time, the site was busy signing up to host a number of external publishers and content companies.

Earlier this year, however, Medium announced that it was pivoting its business away from advertising revenue to a subscription-based model, which reportedly took a number of its publishing partners by surprise and made them rethink their desire to be hosted there.

Vox Media, meanwhile, said that if The Ringer experiment works well, the company may consider opening its platform and network up to other publishers or content creators in the same way. “We may do others, but we will be very selective,” Bankoff told the Journal. “We only want to work with the best and with sites that are consistent with our approach.”

Bankoff, a former executive at AOL, joined what became Vox in 2008. At the time, the company consisted primarily of SB Nation, a network of hundreds of individual blogs written by fans of different local sports teams.

In addition to sites like The Verge, Polygon and Vox, the company also owns Recode, which was previously known as All Things Digital and was at one time co-owned by the Wall Street Journal. In 2015, NBCUniversal invested $200 million in Vox, giving it a theoretical valuation of almost $1 billion at the time.

Court Finds Man Guilty for Liking Defamatory Comments on Facebook

In what appears to be a first, a court in Switzerland has fined a man the equivalent of over $4,000 just for clicking the “like” button on what a judge said were defamatory Facebook comments.

The comments in question suggested that Erwin Kessler, who runs an animal-rights group, holds racist and anti-Semitic views. The defendant (who wasn’t named in the court documents obtained by Agence France Presse) clicked “like” on some of the comments and linked to some of the posts.

Kessler has sued a number of people who participated in those discussions, which began in 2015 during a debate over which animal-rights groups should be allowed to participate in a vegan street festival, according to a Swiss newspaper that covered the story.

Several of the people who made specific comments about Kessler have been found guilty of defamation, but Swiss legal experts said the defendant in the most recent case is the first to have been fined just for “liking” such comments.

According to court documents, the judge in the case ruled that by clicking the “like” button, the defendant “clearly endorsed the unseemly content and made it his own.”

To complicate matters, Kessler was convicted of making racist comments (something that is illegal under Swiss law) in 1998, and briefly served time in prison. But the judge in the recent case said that the defendant had failed to prove that the comments he “liked” were accurate.

Defamation law as it applies to social networks is a grey area in a lot of countries, including the United States, although the U.S. First Amendment provides a lot more protection for an individual’s right to freedom of speech than is found in some other jurisdictions.

Even in the U.S., there have been a number of defamation cases involving social media, including a case in which singer Courtney Love was sued for making derogatory comments about a fashion designer and was forced to pay $350,000 to settle the case.

In Britain, a newspaper columnist was convicted earlier this year of making derogatory comments about a writer on Twitter and was forced to pay damages of $30,000.

The Swiss case, however, appears to be the first in which a man has been found guilty of defamation just for clicking the “like” button on someone else’s comments on Facebook.

In her decision, the Swiss judge argued that by doing so, the man had made the comments “accessible to a large number of people,” since Facebook showed them to all of his friends and followers. Doing this was an “affront to Kessler’s honour,” the judge ruled.

In a Canadian case, a woman was found guilty of making disparaging comments on Facebook about a neighbor, and was held responsible not only for the damage her own comments caused, but also for subsequent comments made by her friends.

Bitcoin Fans Say Crypto-Currency Tokens Are the Future of Tech Funding

Ever since Bitcoin first appeared on the scene several years ago, fans of the crypto-currency have been searching for a way to apply the idea that might capture the public imagination and broaden the use of the technology beyond just geeks and programmers.

Now, some believe that application has appeared with the rise of the “token” economy, in which companies or startup ventures fund their operations by handing out units of crypto-currencies. Some companies have even done what are known as “initial coin offerings” or ICOs, in which they distribute tokens instead of shares to investors.

The crypto-currency market is seen by some as a bubble with hugely inflated prices. Some observers say Bitcoin and other similar ventures are similar to Linux, an open-source alternative to Microsoft’s Windows operating system that has never really achieved mainstream success.

But entrepreneur and investor Balaji Srinivasan, a partner at Silicon Valley venture capital firm Andreessen Horowitz, believes that token-based systems “may eventually create and capture more value than the last generation of Internet companies.”

In an essay published recently on the blogging platform Medium, Srinivasan and his partner Naval Ravikant — co-founder and CEO of a popular online VC community called AngelList — said they believe the token economy has the potential to become “a Kickstarter on steroids.”

The two men, both of whom have been investing in Bitcoin-related technology for several years, argue that using tokens as a financing option has the potential to improve the liquidity options that companies have by several orders of magnitude, as well as increasing the size of the available audience that might want to invest in such ventures.

All of this is possible because of an explosion in the crypto-currency market over the past few years, they argue, in which Bitcoin has survived internal strife but also given birth to alternative currency systems and platforms such as Ethereum.

Initial coin offerings or ICOs are one way of using these new currencies, Srinivasan and Ravikant say. Canadian messaging-app maker Kik recently announced that it is launching its own crypto-currency called Kin, and plans to offer units of it to supporters through a crowdfunding campaign. The currency is based on Ethereum’s blockchain technology.

Kik plans to issue 10 trillion Kin tokens to developers and users via a separate non-profit foundation called the Kin Foundation, which will ultimately hold 60% of all the Kin tokens and be run by a group of independent directors.

Srinivasan and Ravikant warn that some uses of crypto-currency tokens, including some ICOs, may be subject to regulation by governments, if they are seen as equivalent to doing a traditional equity offering or IPO, in which investors receive shares of the company. But they argue other uses of tokens for crowdfunding could essentially be unregulated.

Token supporters say they aren’t really equity but more of a digital IOU, which entitles the holder to redeem their tokens in return for access to a platform like Ethereum’s.

That access has value because it can be used to generate Bitcoin-style coins through a computer-intensive process known as “mining,” and those coins can in turn be exchanged for other more familiar currencies like U.S. dollars. One Bitcoin is currently worth about $2,300 U.S.

Some skeptics say token-based fundraising has the potential to turn into a huge boondoggle if it is unregulated, with unwary investors being fleeced of their savings with little to show for it.

Ravikant and Srinivasan, however, argue that tokens will allow companies to raise money much more quickly for new ventures than existing systems do, and will also allow for startups to build valuable services without having to rely on advertising as their only revenue source.

Large technology companies like Google and Facebook offer “have sometimes come under fire for making billions of dollars while early adopters only receive the free service,” their essay says. “After the early kinks are worked out, the token launch model will provide a technically feasible way for tech companies to spread the wealth and align their user base behind their success.”

Robotics Expert Says Tesla’s Self-Driving Tech Is a Danger to Cyclists

Tesla’s autonomous or self-driving car technology has helped make the company a **-billion success story. But the system does have flaws, according to a Stanford University robotics researcher, and one of them is a disturbing inability to recognize cyclists.

Heather Knight is an expert in human-robot interfaces who has a PhD from the Robotics Institute at Carnegie Mellon University and is working on her post-doctoral research in social robotics. She published her concerns about Tesla’s self-driving tech in an essay on the blogging platform Medium, entitled “Tesla Autopilot Review: Bikers Will Die.”

Knight described how she and a colleague from Stanford took a Tesla for a drive in order to see how its self-driving technology matched their expectations about human-robot interaction.

Although Knight said the Tesla vehicle’s autopilot worked flawlessly in most situations while the two drove around the streets and highways of southern California, she said she was “concerned that some will ignore its limitations and put biker lives at risk; we found the Autopilot’s agnostic behavior around bicyclists to be frightening.”

In particular, Knight said that the Tesla’s “situation awareness display”—which helps human drivers understand what the car’s self-driving technology can “see” in the environment around it—was fairly accurate with cars, but not even close to accurate with cyclists.

Although she gave the display an A+ rating overall, the Stanford researcher said: “I’d estimate that Autopilot classified ~30% of other cars, and 1% of bicyclists. Not being able to classify objects doesn’t mean the tesla doesn’t see that something is there, but given the lives at stake, we recommend that people NEVER USE TESLA AUTOPILOT AROUND BICYCLISTS!”

There appear to have been no reported incidents where Tesla cars driving on auto-pilot came into contact with cyclists, although there was an accident in Norway in 2016 where authorities said a Tesla failed to properly recognize a motorcycle and injured the rider. In a Wall Street Journal article, one Tesla driver credited the autopilot with preventing an accident with a cyclist.

In a number of responses to readers of her piece, both on Medium and on Twitter, Knight said that she likes Tesla, but was trying to point out that its system is still what she called a “human in the loop” technology because of such inaccuracies in detection.

Tesla makes it clear to drivers that its autopilot system is not fully autonomous, and that they should keep their hands on the steering wheel and pay attention to their surroundings at all times. If a driver takes their hands off the wheel for a sustained period of time, the autopilot disengages and can’t be re-enabled unless the car is stopped.

Despite its flaws, Knight said that the Tesla’s Situation Awareness Display was her favorite feature of the car, because “it helps the driver understand shortcomings of the car, i.e., its perception sucks.” Robot-based systems in general, she added, “would benefit from communicating their limitations to people.”

Some of those who criticized Knight’s essay said that it didn’t provide enough evidence of flaws in the detection system, and that the headline on the article was unnecessarily inflammatory about the risks that autonomous driving poses to cyclists.

Knight said that her concern is that people will see the so-called “autopilot” feature as being fully autonomous at all times, and will fail to pay attention to their surroundings, “as we see in the fatal crashes so far.” Even if detection rates improve, she said, it’s important that human drivers “have the correct mental model of the car, complete with its shortcomings.”

Fortune has asked Tesla for a comment on Knight’s conclusions, and will update this post if and when one is provided.

Here’s Why Facebook Opened Instant Articles up to Google and Apple

If there’s one thing you can usually count on with Facebook, it’s that any new feature the giant social network introduces will be designed primarily to benefit Facebook, in most cases by getting users to spend more time inside the company’s walled garden.

That’s why an announcement Facebook made on Thursday is unusual. The company said it is opening up the technology behind its mobile-focused Instant Articles feature so it works with similar services offered by Google and Apple.

With one step, publishers will now be able to use a Facebook software tool to produce articles that fit the Facebook IA standard, and also comply with Google’s competing AMP (Accelerated Mobile Pages) standard and the Apple News format.

This doesn’t mean Google or Apple will get any more access to Facebook’s platform than they would have before, but it does represent a kind of truce in the mobile news war, and in that sense it is a tacit admission of failure by Facebook.

By supporting Google and Apple’s technologies, the company is essentially admitting that Instant Articles by itself wasn’t enough to entice publishers.

For the past two years or so, ever since it launched Instant Articles as a limited trial with partners like the New York Times (and Time Inc., which owns Fortune), the social network has been trying to get media companies to adopt it.

In a nutshell, the feature takes articles that have been formatted for the web and strips them down to make them load more quickly in Facebook’s mobile browser.

Initially, this was an appealing idea for many publishers, who either didn’t have the financial resources or the skills to make their stories load faster in mobile browsers. But at the same time, it was a problematic deal in many ways.

Facebook offered anyone who participated in the program 70% of the revenue from ads that Facebook was able to sell, or 100% of the revenue from ads they sold themselves. But this revenue didn’t exactly start pouring in for many of those who signed up.

The company made a number of tweaks to try and improve the situation, but it didn’t really move the needle for many media companies, especially since one of the main ways Facebook sped up the loading of pages was by blocking certain types of ads.

In addition to that, some publishers were leery of giving Facebook even more control over their content. The giant social network already accounts for a huge amount of the traffic that many media outlets get from the web, and uses its algorithm to control who sees their content and when.

Some, like the Washington Post, jumped in with both feet, but others — including Bloomberg, the Wall Street Journal and ESPN — balked.

Even the New York Times, a launch partner, eventually stopped participating in Instant Articles, and so did The Guardian (which also pulled out of Apple News). In part, that’s because both papers were focusing on their digital subscriptions and membership programs, and Instant Articles wasn’t really helping.

Many publishers seem to be more interested in Google’s AMP standard, which is more open than Facebook’s. Although the search giant is in charge of the technology, it is an open-source project that makes its code public and theoretically allows anyone to take part in developing it.

There are still concerns on the part of some media companies that Google has too much control over AMP, and that it is interested primarily in promoting its own advertising networks and technologies, but AMP still seems to have gotten more uptake than Instant Articles. It also supports subscriptions and paywalls, which Instant Articles doesn’t.

Apple News, meanwhile, got off to a slow start, with many publishers seeing extremely low traffic from articles they formatted for the service. But more recently there is evidence that some sites are seeing improved traffic from their Apple News articles.

Apple also recently hired a managing editor for Apple News, a position that didn’t exist before, which suggests that it might be looking to beef up the service and possibly even expand it.

Facebook’s New Video Deals Show How Desperately it Wants to be YouTube

Facebook may be the world’s largest social-media platform, but there is one significant place where it still falls short, and that is video. YouTube still holds the record in that category, with more than 1 billion hours of video watched every day — 10 times what Facebook generates.

The giant social network didn’t get where it is by settling for second place in any market, however, and that has meant an unprecedented push by CEO Mark Zuckerberg into video in all forms.

The strategy began with a focus on streaming via Facebook Live, which launched in **, a feature that was aimed originally at celebrities and then expanded to include paying for live-streamed content from traditional media entities like the New York Times.

More recently, Facebook has moved aggressively into buying, commissioning and licensing longer-form, more TV-style content.

There have been reports for some time that Ricky Van Veen — the CollegeHumor co-founder Facebook hired last year — was looking to sign deals with a wide range of media companies, TV networks and individual artists for anything video related.

On Wednesday, some details of those arrangements finally came to light, when Reuters reported that the company has committed to distribute both short-form and longer-form video programming from a number of players — including BuzzFeed, Vox Media, ATTN and Group Nine Media, a New York-based outfit that owns short-form video producer NowThis.

Just as it did with its live-streaming efforts, Facebook has shown it is willing to pay up for the video it is getting from its various partners.

According to Reuters’ sources, the company is going to pay between $10,000 and $35,000 for videos that run 5 to 10 minutes long, and as much as $250,000 for any video that is 20 to 30 minutes long. The latter will be owned by Facebook, according to the wire-service report, while the creators of shorter clips will keep all the associated ownership rights.

Judging by the description, the shorter videos Facebook has in mind will be very much like the content that providers like ATTN, NowThis and BuzzFeed — with its popular Tasty cooking clips — already create for the social network.

The longer videos are a different beast, however, and they mark a distinct change in Facebook’s strategy. Instead of just being ad hoc clips of something newsworthy or funny designed to get clicks, the social network appears to have its sights set on what are called “scripted” shows, which is much more like what YouTube and others do.

YouTube has always carried episodic shows and scripted content, but it made a significant push into this as a business with the launch of a subscription service called YouTube Red last year.

As part of that roll-out, the Google subsidiary signed deals with a number of creators, some of whom got their start on the platform — including controversial game reviewer PewDiePie — as well as more mainstream sources such as ** and **.

Since then, the field of TV-style digital video has gotten even more crowded. Snap, which has been trying to justify its $20-billion market cap ever since it went public in March, has also been signing up a variety of players to create shows for its video messaging app. Partners in that effort include late-night TV host James Corden and NBCUniversal.

Mark Zuckerberg has said that he sees longer-form, scripted content as an “anchor” for a new video tab on the site, and that this will be a destination for those looking for video to watch. In other words, exactly the same position that YouTube currently occupies for many younger users of the web.

As with Snapchat, however, the multi-billion-dollar unanswered question is: Do people really want to watch longer-form, TV-style content on Facebook?

In much the same way that Snap is (and Twitter as well), Facebook seems to be moving towards video not because users want it, but because video is the place where it can get the most money from advertisers. Engagement levels are higher with video, meaning people tend to spend more time with them, and therefore they are seen as being more valuable.

As the traditional TV universe continues to fragment and disintegrate due to cord-cutting and other forces, Facebook and others like Snap see the potential shift of billions of dollars in TV ads, and they very much want to be the place where that money lands.

Among the hurdles that remain are a) coming up with content worth watching, and b) proving to advertisers that social views are as valuable as TV views. For the purposes of advertising, Facebook counts as a view anything that lasts longer than three seconds.

The comparison between social media and TV has been a bone of contention for some time in TV land, with a number of traditional networks arguing in their recent “Upfront” presentations to advertisers that television is worth more because it reaches real people. “**,” said NBC’s **.

In Facebook’s case, the issue is complicated by the fact that the social network has repeatedly had to admit errors in its audience-measurement analytics, including one mistake it admitted to last year, in which it over-estimated video views for more than two years.

Snapchat Tries to Reinvent TV For Mobile-Addicted Millennials

As Snap tries to outrun competitors like Instagram, which is busy copying every feature it has, the company behind the popular Snapchat messaging app is having to keep pushing the goalposts further down the field. Its latest big push is TV-style video.

Snap has been experimenting with short-form video “shows” for some time now, including some one-off projects related to Saturday Night Live as part of a deal with NBCUniversal, which recently invested $500 million in the company. But now it seems to making a major effort to roll out the strategy with a series of high-profile announcements.

Late-night talk show host James Corden is doing a show for the app that involves a talent search for the next great late-night host, and comedian Conan O’Brien is producing an animated comedy show. NBC, meanwhile, is slated to deliver a daily news show later this year.

Snap is also working on a multiple-show deal with the Discovery Network, one that will begin with a “Shark Week” short. According to a report in Variety, Snap wants to quadruple the number of shows it has from 5 to 20 by the end of the year.

The company says that its push further into TV-style content came as a result of seeing the success of its first such effort: A news show called Good Luck America that was introduced first during the election campaign, starring Snap’s head of editorial and former CNN reporter Peter Hamby.

Snap says the show got 5.2 million viewers per episode in its latest season, although it should be noted that a “viewer” on Snapchat isn’t the same as a viewer on television. Snap counts as a view anyone who clicked to start a video.

In a larger sense, Snap’s move into video is not surprising. Video, and particularly mobile video, is one of the few areas in the digital-ad market that is showing some strength, and almost every major technology and/or media company from Facebook to Amazon is going after it with guns blazing.

Facebook, for example, has made it clear that it wants to do deals with anyone who has short-form TV-style content, and it is willing to pay up in order to get the rights.

The motivation for Snap to be a part of this video land-rush is even more compelling than it is for Facebook or Amazon or YouTube. After an initial public offering in March that gave it a theoretical market value of $27 billion, Snap is facing a considerable amount of pressure to show that it can increase revenues, and quickly.

Snap CEO Evan Spiegel likes to boast about the amount of time users spend with the Snapchat app, but many analysts and investors seem concerned about things like the declining rate of user growth and the risk that advertising revenue may not match expectations. That’s what helped shave $6 billion off the company’s market cap following its recent earnings report.

One way to address this problem is to give advertisers and investors something they understand, which is TV-style content. But the big unanswered question is: Will Snapchat users take to this format? Or do they mostly just want to exchange personal messages with their friends?

In a way, all the things that have made Snapchat such a viral hit over the past couple of years are the same things that could make it difficult for it to succeed with a larger mass-media strategy.

The key appeal for many users in the early days was that Snapchat messages only went to a small group of friends, and they disappeared within 10 seconds. This made it perfect for intensely personal messages. And Snapchat played to this use case by not providing the usual social-status indicators like follower counts and view counts and so on.

As it has grown larger and felt the need to justify its eye-popping valuation, Snap has tried to build a much broader media strategy on top of this focus on personal messaging. It started with Discover, a feature that added “stories” from news entities like CNN and BuzzFeed.

Video is the next iteration of that strategy (Snapchat’s TV shows will have their own section within the Discover tab, Variety says). But will this effort prove that Snap can combine its millennial audience with the revenue-generating ability of video? Or will it take the company too far away from its roots?

As we’ve seen with Instagram’s relentless copy-cat approach over the past year, another big risk for Snap is that the Facebook-backed company — or even Facebook itself — will simply move in on its market even further and suck all of the oxygen out of the room when it comes to video.

Snap will undoubtedly point to its millennial reach and those much-talked-about engagement numbers as a way of arguing that it is more valuable. And that might work for some advertisers.

At the same time, however, Facebook’s 1.8 billion users are a powerful draw, and so are the massive amounts of money that the social networking giant has the ability to spend on TV-style content. And even if Facebook doesn’t prove to be the Snap-killer when it comes to video, there’s always Amazon and YouTube and likely Apple as well to contend with.

No one said being a $20-billion startup was going to be easy. Welcome to the jungle, Evan.

People Distrust the Media in General, but Trust the Media They Like

Studies have found that trust in the media is not only extremely low, but falling. A new report from the American Press Institute, however, suggests that the answer to surveys about trust and the media can change a lot depending on how the questions are phrased.

The study was done as part of the Media Insight Project, which is a collaborative effort between the API and the Associated Press-NORC Center for Public Affairs Research. It appears to show that public attitudes about the media are “more complex and nuanced than many traditional studies indicate,” according to a release from the Institute.

In particular, people’s answers changed depending on whether the question was about their perception of the media in general or the media that they used most often—which in most cases consisted of fairly mainstream media outlets.

So when people were asked whether the media in general were normally “very accurate,” only 17% said that they agreed with the statement. But when respondents were asked about the media sources that they rely on the most, twice as many—34%—said they believed they were very accurate.

In a similar way, only 22% of people said that the news media in general cares about the people it reports on. But more than 35% of those surveyed said that the outlets they rely on care.

Interestingly enough, the study showed that divisions between Republican voters and Democratic voters when it comes to trust in the media virtually disappear in some cases when the question refers to media sources that the user relies on most.

So while only 8% of Republicans said that the news media in general was “very accurate,” 40% of Republican voters said the media sources they use most were accurate, roughly the same as the number of Democrats who agreed with that statement.

Many recent studies have highlighted an ideological divide when it comes to trust in the media. A report from PR strategy firm Edelman called The Trust Barometer, for example, found that only 15% of Trump voters said they trusted the media to do what was right, compared with 51% of Clinton voters who agreed with the same statement.

According to the American Press Institute, the findings from its study suggest that the issue of trust in the media “is more complicated than some may think.”

In particular, the API says the research shows that the idea of Americans somehow retreating from news in general, or being separated into their ideological corners “oversimplifies what is occurring.” However, the Institute did note that trust in the media among those under 40 does appear to be declining, regardless of ideology.