Weeks after NASA's New Horizons spacecraft made the farthest-ever visit to an object, located more than 4 billion miles from Earth, the probe beamed home an unprecedented photo.

The image (above) is devoid of color and looks somewhat fuzzy, but it represents the most detailed look yet at the object. Researchers say there are more and even better pictures that have yet to arrive.

Early in the morning on New Year's Day, NASA's nuclear-powered mission zoomed past the snowman-shaped object, which is formally known as (486958) 2014 MU69, though it's commonly called "Ultima Thule" (see the editor's note below).

New Horizons flew within 2,200 miles of MU69 at a speed of about 32,200 mph, and data from the flyby is giving scientists a window into the formation of the solar system. Since Mu69 is so cold and far away, it's essentially the most pristine building block for planets ever seen.

"This is exactly what we need to move the modeling work on planetary formation forward, because we're seeing evidence — right here — of accreting objects, and then having them combine," Cathy Olkin, a deputy project scientist on the New Horizons mission, said during a press conference earlier this month.

Read more: NASA's deep-space nuclear-power crisis may soon end, thanks to a clever new robot in Tennessee

However, as Olkin and researchers have known for a long time, it will take about 20 months — likely into late 2020 — to download all of the data from New Horizons' historic maneuver.

That's why researchers released the "best-yet view" of Mu69, shown above, just this Thursday — more than three weeks after the flyby.

In the comparison below, the left image is one of the earliest photos of the object sent to Earth, photographed by New Horizons from about 85,000 miles away. On the right is the new image, which was taken from about 4,200 miles away. Both are sharpened to show more detail.

An early view of MU69 (left) vs. the most detailed version yet (right).NASA/JHUAPL/SwRI

The new photo is far from the last that researchers will get access to, however; New Horizons took hundreds of images and recorded scores of other data during the flyby.

"Over the next month there will be better color and better resolution images that we hope will help unravel the many mysteries of Ultima Thule," Alan Stern, who leads the New Horizons mission, said in a press release published on Thursday.

An illustration of the Kuiper Belt with New Horizons' flight path, Pluto, and Ultima Thule (or 2014 MU69).NASA/JHUAPL/SwRI/Alex Parker

When New Horizons flew past Pluto in July 2015— the first-ever visit to the minor planet — it took the probe about 15 months to send back all of its data to Earth. In the case of Mu69, it will take about 20 months to send everything it recorded on New Year's Day.

The wait is so long because of the hardware on and distant location of New Horizons.

NASA built its probe in the early 2000s, then launched it from Earth in 2006. So while New Horizons' systems are redundant and healthy, the electronics themselves are outdated by about 15 to 20 years.

New Horizons is featured on a 2016 US Post Office stamp. USPS Another bottleneck is the weak signal from the spacecraft's radio antenna. Right now, its power output is about 15 watts, or one-fourth that of a standard 60-watt incandescent light bulb.

A third problem is that it's broadcasting from more than 4 billion miles away. At this distance, each bit of digital data — sent as radio waves traveling at light-speed — takes more than six hours to reach antennas on Earth.

Read more: The speed of light is torturously slow, and these 3 simple animations by a scientist at NASA prove it

These factors throttle the probe's output to below 1,200 bits per second, which is about 80,000 times slower than the average broadband internet download speed in the US in 2018.

At this rate, it took weeks to get enough images to create a small movie showing the rotation of MU69, which spins about once every 16 hours.

The New Horizons team expects to see the highest-resolution color photos sometime in February.

"We are guardedly optimistic that those highest-res images will cover a massive amount of the surface," Stern said earlier this month. "Stay tuned for February."

Planetary scientists like Stern are eager to see more because the pictures and other data should help solve some longstanding mysteries about the solar system's 4.5 billion years of history.

"It's like the first time someone opened up the pharaoh's tomb and went inside, and you see what the culture was like 1,000 years ago," he said. "Except this is exploring the dawn of the solar system."

Editor's note: After a public campaign, the New Horizons team selected Ultima Thule as a nickname for (486958) 2014 MU69. However, we've de-emphasized the nickname because the Nazi party used the word "Thule" as a tenet of its ideology.

Kanye West may have said it best in his song "Power" when he sings, "no one man should have all that power." That's where these prominent, high-profile couples come in.

While some tech leaders like Mark Zuckerberg have been with their partners since college, some notable figures in the tech sector have gravitated toward partners with just as much power and pull in their industries.

There have also been some high-profile splits in the last year that have hit some prominent tech leaders. Amazon CEO Jeff Bezos and his wife, MacKenzie, announced in early January that they were getting a divorce. Tesla's CEO Elon Musk and Canadian singer Grimes dated for a while, though it's unclear whether the on-again off-again couple was still together come 2019.

Here are 13 of the top power couples in the tech industry:

Penalties from California regulators are raising new questions about workplace safety at Tesla's vehicle production facilities in Fremont.

California's Division of Occupational Safety and Health (Cal-OSHA) inspected GA4, an assembly line Tesla constructed in an open-air structure outside its main factory in less than a month, between June 21 and December 18 (the agency did not specify the number of times it visited the facility) and issued penalties for six violations of California's labor regulations. The citations resulted in fines that total $29,365.

According to Cal-OSHA, Tesla failed to do the following:

Tesla appealed the citations, saying for each that it did not violate the specified regulation and that the corresponding penalty was unreasonable.

"Nothing is more important to me or to Tesla than the health and well-being of our employees," Laurie Shelby, Tesla's vice president of environmental, health, and safety, said in a statement to Business Insider. "My EHS team and operational leaders have been intently focused on GA4 over the past six months, implementing safety protocols throughout the new line that not only keep Tesla in compliance with existing standards but also reduce risks to associates. The OSHA inspection did not result from any incident or injury and occurred during the construction phase of the project. Tesla will challenge OSHA's findings regarding the safety conditions that were present at the time of GA4 construction,"

Read more: Ex-Tesla employees describe the abrupt way they were laid off and say questions linger

Thomas Armstrong, a professor at the University of Michigan who has worked with auto companies to make their manufacturing processes more ergonomically friendly, told Business Insider that the violations are not surprising due to the speed with which Tesla built GA4, but he said the violations do not necessarily indicate that the facility is unsafe, since they largely refer to failures to take preventive measures, rather than worker injuries.

"It certainly is not an endorsement for their safety culture, but, again, it doesn't mean that they aren't operating safely," he said.

The kinds of violations Tesla was cited for are common in the auto industry, and the fine Tesla received amounts to "a slap on the wrist," Armstrong said.

Questions about worker safety

Injury statistics and reports from media outlets have raised questions about worker safety at Tesla's factories, though concerns about workplace safety are not unique to Tesla in the auto industry.

A 2017 report from the worker advocacy group Worksafe said the injury rate at Tesla's Fremont factory was 31% higher than the industry average in 2015 and 2016, but Tesla said in 2018 that it had made safety improvements that led its recordable injury rate to fall by nearly 25% in 2017. The automaker said its 2017 recordable injury rate was "equivalent" to the industry average.

Worksafe has not released an analysis of Tesla's 2017 or 2018 injury rates, but Tesla received more citations from OSHA related to vehicle manufacturing than Ford, General Motors, or Fiat Chrysler from 2017 through the end of 2018. Tesla received 21 citations during that period, while Ford received 15, General Motors received three, and Fiat Chrysler received one. Ford was fined the most during that period, $102,554, while Tesla was fined $63,870, General Motors was fined $14,122, and Fiat Chrysler was fined $7,967.

While OSHA's online database does not disclose the nature of each violation, reports from Reveal published in 2018 claimed that Tesla misreported workplace injuries, avoided using safety markings for aesthetic reasons, and failed to give injured employees proper medical care. The first report was published before Tesla built GA4, and the second made no specific mention of the facility.

Tesla has denied that it has misreported workplace injuries and failed to use safety markings for aesthetic reasons. The automaker did not respond to requests for comment on the allegation that it failed to give injured employees proper medical treatment.

An unconventional decision

The decision to build an open-air structure to supplement vehicle production in less than a month was unconventional. It was motivated by a desire to hit production goals for the Model 3 sedan, which had been subject to significant production delays in the prior year.

Michael Ramsey, an automotive analyst at Gartner, told Business Insider at the time that the move had no precedent in the auto industry.

"It's extremely unusual, at a minimum," he said. "I've never heard of anyone ever doing this before on a grand scale."

But Tesla CEO Elon Musk praised GA4 on multiple occasions, saying he preferred it to the main factory and highlighting how it increased the automaker's production output by 50%.

"It's actually way better than the factory building. More comfortable & a great view of the mountains," Musk said in June.

Have a Tesla news tip? Contact this reporter at This email address is being protected from spambots. You need JavaScript enabled to view it..

Airbnb has acquired Gaest.com, an online marketplace for renting out meeting spaces for just hours at a time, the company announced Friday.

Gaest was founded in Denmark in 2015 by CEO Anders Mogensen. It's used by guests to book spaces for meetings, team building events and "even photoshoots," according to the press release. Gaest lets hosts list their own spaces for rent just like Airbnb lets hosts rent out their personal spaces to vacationers.

The terms of the deal were not disclosed. Airbnb was last valued at $31 billion in 2017.

"Working professionals may spend more than half their waking hours at work. Meetings and events offer a unique opportunity for Airbnb to address local use cases and support our full community of diverse professionals," Airbnb said in a statement.

Gaest.com, which offers it service in numerous countries across six continents, will operate as a unique platform for now, though its employees will join Airbnb, according to the statement.

The acquisition comes just a couple of weeks after Airbnb disclosed that it's been profitable for the last two years, at a time when profitability is quite rare for even the most successful venture-back startups, including Uber and Lyft.

Airbnb is also rumored to be pursing an IPO.

It also highlights a new growth strategy with Airbnb. By moving into workspace, Airbnb gears up to compete against companies like WeWork, which let nomadic workforces rent out desk space and meeting rooms for short periods of time.

WeWork, which recently rebranded to The We Company, was last valued at $47 billion in a January financing round led by SoftBank.

You've likely heard that learning a new language is easier for children than it is for adults. But all hope is not lost for grown-ups looking to expand their linguistic knowledge.

While a recent MIT study did pinpoint 17.4 years as the cut-off for obtaining a native-like mastery of another language's grammar structure, the results of the same study suggest that people over the age of eighteen can definitely still learn a foreign language quickly - they just may not achieve the same native-like proficiency.

It is also important to remember that children are not inherently better learners than their adult counterparts: the two groups just learn very differently, and, in fact, there are a few ways that adults can outperform children in language acquisition based on their longer and more robust life experiences.

If you're a person over the age of eighteen determined to learn a foreign language, read on for five easy strategies you can employ today:

Condé Nast said this week that it would put all its titles behind a paywall by the end of the year and that it would charge marketers a premium to advertise to digital subscribers.

Ad buyers expressed doubts that the glossy-magazine publisher's paywalled audience is worth paying more for, though.

Chris Mitchell, chief business officer for Condé Nast's culture division — which includes The New Yorker, Wired, and Vanity Fair — said the company planned to start charging a premium probably in the second quarter. That premium would apply across the aggregated digital subscriber audiences of those three titles, which already have paywalls in place that people hit after reading four articles a month.

Read more: Glamour was selling millions of magazines, but still couldn't make the business work. And it shows the dire state of advertising.

Currently, Wired's digital subscribers get an ad-free experience, but that would change. Ars Technica, Condé Nast's specialty tech publication, also could be part of the experiment.

Subscribers are more engaged than non-subscribers

Making the case for charging a premium, Condé Nast said that The New Yorker subscribers spend on average 7.3 times as much time on the site per month as non-subscribers do.

On Wired, subscribers spend 8.7 times as much time, and for Vanity Fair, it's nine times as much.

"We know they're more engaged, they've visited more articles to be behind the paywall, they're visiting more articles than the people who visit through search," Mitchell said. "So they're inherently more valuable."

Publishers of all kinds have been turning to paywalls and membership programs to prop up their businesses. Monica Ray, Condé Nast's EVP of consumer marketing, said the rest of Condé Nast's titles, including Vogue, GQ, and Bon Appétit would go behind the paywall starting in the third quarter.

"I think it's going to be a tough sell," said John Wagner, group director of published media at PHD, of the idea to charge a premium for paywalled readers.

Wagner said Condé Nast will have a hard time selling this way because other paywalled publications don't sell this way. And advertisers could still try to buy around the premium by buying programmatically.

The New York Times and The Wall Street Journal, for example, don't sell their audiences this way.

Advertisers would also expect to get a discount for the non-paywalled audience if the publisher charges more to reach the paywalled audience, he said.

Ad buyers are pushing back

Barry Lowenthal, CEO of The Media Kitchen, challenged the idea that people who pay for content are more valuable than freeloaders.

"I don't think it's any more valuable then free content," he said. "Of course they will try and say that subscribers are more valuable — maybe richer and more engaged — than non-paying readers, but publishers have spent decades telling us that it doesn't matter. People may also assume that if they pay for content, they won't see ads."

Mitchell pointed out that Condé Nast has always maintained that most of its print copies are paid for — it's just trying to test the notion of applying that same thinking to the digital sphere.

As for lowering the rates for non-paywalled audiences, he said: "I don't think there's anything wrong with the current CPM; this isn't about lowering the value of the user base, it's about identifying the higher value of people behind the paywall."

Still to be worked out are things like how much of a premium will be charged, and if the ad experience behind the paywall will be different from the one in front of it. Condé Nast will pay attention to how the ad market responds to the pricing change and how digital subscribers engage with ads compared to non-subscribers, Mitchell said.

Condé Nast has tried before to charge advertisers more to reach audiences based on their path to the sites, based on the idea that people who come from Facebook, for example, stay more on site than people who come from Twitter. The practice hasn't caught on widely, though.

Condé Nast will experiment with dynamic paywalls

Paywalls can impact traffic and, in turn, advertising revenue. Ray said the paywall approach would vary by title but that the plan is that most people wouldn't hit the paywall, so there would be minimal impact on traffic.

"For a lot of longform journalism, the meter makes sense," she said. "For brands like Bon Appétit, there's so much engagement around recipes; Architectural Digest, there's so much engagement around its photo archive. We're really going after this longtail audience."

Ultimately, the company might try other value exchanges for people who hit the paywall but are resistant to subscribing. Once a reader hits the paywall, they might get another free article that's sponsored by an advertiser, or get to read an article after watching an advertiser's video, for example.

Ray also said the company plans to experiment with so-called dynamic paywalls, as The Wall Street Journal and The New York Times use. That would let Condé Nast vary the number of free articles someone is able to see depending on their propensity to subscribe.

One of Facebook's most high-profile executives has been flouting the social network's rules on using your own name on your profile for eight years.

Andrew Bosworth, the company's head of AR and VR (augmented and virtual reality) and a key lieutenant of CEO Mark Zuckerberg, is commonly known as "Boz."His personal Facebook page lists his name only as "Boz," in violation on Facebook's controversial rules around names, Business Insider has found.

The California social network has been repeatedly criticized for its rules that stipulate that users must use their real names on the platform. In 2014, executive Chris Cox was forced to apologize to transgender and other LGBTQ Facebook users after the policy was abused to target them.

But Bosworth's rule-breaking suggests that the company is less worried about enforcing its rules when it comes to its own leadership team.

Reached for comment, Facebook spokesperson Tera Randall said that it was a "mistake" and Bosworth's name is going to be changed to "Andrew Bosworth (Boz)."

Andrew Bosworth's Facebook profile, which lists his name only as "Boz". Facebook

The rules, and how 'Boz' breaks them

The current real-name rule, according to Facebook's public help page, reads: "The name on your profile should be the name that your friends call you in everyday life. This name should also appear on an ID or document from our ID list," a list that includes official government documentation like passports and driver's licenses.

Unless Bosworth has legally changed his name to simply "Boz," then "Boz" alone breaks this rule.

The help page then adds: "Nicknames can be used as a first or middle name if they're a variation of your authentic name (like Bob instead of Robert)."

"Boz" would clearly qualify as a nickname — but Bosworth isn't using it merely as a first or middle name. It's his entire name on his Facebook profile, without a last name, meaning he is still in violation of the rules.

Business Insider attempted to remove this reporter's last name on Facebook, to see if the social network would allow users to go by their first name (or a nickname) only. It would not, displaying an error message saying: "You must provide a last name."

Additionally, users can list "another name" on their account, including "maiden name, nickname, [or] professional name." But doing this normally lists the additional name alongside their actual name — not instead of it.

Bosworth's profile has also been verified by Facebook, indicating that the social network has confirmed that it really belongs to him and isn't an impostor.

Facebook doesn't allow ordinary users not to have a last name. Facebook

The difference between Pages and Profiles

There are two different kinds of user-pages that users can create on Facebook: "Pages," and "Profiles."

A profile is a personal account directly linked to you, showing your real name (according to Facebook's rules, at least). Users need a profile on Facebook to do almost everything, from making comments to adding friends. Content posted to profiles can be either public or private, at the profile owner's discretion, and if the profile-owner allows it other users can "follow" their profile, meaning their public posts show up in the followers' news feeds.

A page, meanwhile, is a public page users can make about themselves, or a business they own, or things they like, or just about everything. Other users can then like these pages and receive posts from them their News Feeds. There are no rules about what pages can be called, beyond the standard ones around obscenity or illegality.

A Facebook exec (or anyone else) could make a page using a fake name and then post from it without breaking any rules. But this isn't what Bosworth is doing. He is using the "Boz" name on a personal Facebook profile.

Bosworth has used the same profile to make both personal and professional posts about his life for more than a decade. He makes public posts from it to this day, and it now has nearly 125,000 followers on it — but it's still a profile.

At least one other Facebook executive who is widely known by a nickname internally uses their real name on their Facebook profile. Kang-Xing Jin is often referred to as KX, but his name is listed on his profile as "Kang-Xing Jin."

A post on Andrew "Boz" Bosworth's profile from 2006. Facebook

It's not the first time Facebook has employed double-standards for its executives

Facebook's "discriminatory" real-name policy has caused huge upset among users over the years, from trans users who feel unable to express their true identities on the platform to users from ethnic groups whose names don't sound "real" to Facebook's moderation team.

The Silicon Valley tech giant has defended the rule as necessary to ensure users know who they're really communicating with, and to protect people from harassment, scams, and anonymous abuse - even if it inadvertently causes pain to users penalized by the policy.

But the company seems much less interested in enforcing the rule when one of it is own executives violating it.

In a statement, Facebook's Randall said: "Boz has had this profile name since 2011 when one of our automated systems from that time prompted him to make the change due to what our technology and policies allowed at that point. Since then we've instituted our ID-name match requirement ... We never changed it when the policy changed, which was a mistake we will correct by returning it to Andrew Bosworth (Boz). He has never asked for an exception to our policies or used our systems in any way not available to people at the time."

The apparent double-standard is reminiscent of an incident in April 2018, when tech news site TechCrunch discovered Facebook had secretly been deleting messages Zuckerberg had sent other users, without the recipients' knowledge or consent. The company claimed it was necessary for "corporate security" and hastily promised to roll out an "unsend" feature for all users — but nine months later the feature still appears to be in the testing phase.

Facebook exec Andrew "Boz" Bosworth gives a talk at the Online Marketing Rockstars marketing trade show in Hamburg, Germany, 03 March 2017. Christian Charisius/picture alliance via Getty Images

Andrew Bosworth is a key adviser to Mark Zuckerberg

Andrew Bosworth first joined Facebook in 2006, and is a member of of the "M-Team"— an elite group of execs that help steer the company and advise Zuckerberg. He has held a number of roles over the years, including helping launch News Feed and leading the company's advertising efforts, and is currently overseeing Facebook's AR and VR efforts as VP of AR/VR.

In 2018, BuzzFeed News published an internal memo written by Bosworth from 2016 in which he defended Facebook's growth at any cost, even if people died as a result.

"The ugly truth is that we believe in connecting people so deeply that anything that allows us to connect more people more often is *de facto* good," he wrote. "It is perhaps the only area where the metrics do tell the true story as far as we are concerned."

Bosowrth subsequently said he doesn't agree with the memo, and didn't believe what he wrote even when he wrote it.

Do you work at Facebook? Got a tip? Contact this reporter via Signal or WhatsApp at +1 (650) 636-6268 using a non-work phone, email at This email address is being protected from spambots. You need JavaScript enabled to view it., Telegram or WeChat at robaeprice, or Twitter DM at @robaeprice. (PR pitches by email only, please.) You can also contact Business Insider securely via SecureDrop.

DAVOS — Tech will never be the same again after 2018. That's the view of Brad Smith, Microsoft's president and chief legal officer, after a "watershed" year of controversy and regulation for the sector.

Referring to the so-called "tech lash," which was in part sparked by Facebook's giant Cambridge Analytica data breach in March 2018, Smith said the incident "captivated people's attention in a way that data breaches or issues in the past had not."

The European Union significantly strengthened privacy laws with the introduction of GDPR, while California broke ground with its own privacy legislation last year. The former is already showing its teeth, with Google being hit with a $57 million fine by French authorities for its "massive and intrusive" data collection.

"It was one of these watershed years where I doubt that we'll ever go back to what was there before. People now recognise that the impact of technology is so pervasive that it requires that new steps be taken to address what's on people's minds," Smith told Business Insider at the World Economic Forum in Davos, Switzerland.

Some of these steps should come from within the industry, Smith said, others will involve government input. The likes of Apple, Facebook, and Google all support the introduction of federal privacy laws in the US, and Smith thinks they could be put in place as early as this year.

"It is possible we will see a law passed in 2019," he said, pointing out that Microsoft has lobbied for new legislation since 2005. "It's more likely that we'll see a law passed in 2020 or 21. It is now a historical inevitability that a law will be passed by Congress."

Read more: The president of Microsoft says top leaders at the world's biggest tech companies meet regularly to talk about the major issues facing Silicon Valley and the world

On how regulation will shake out, he said GDPR will be a useful starting point — not least because tech firms won't want to grapple with two substantially different sets of rules on managing user data.

It means US tech giants, and others guilty of abusing data, could face big financial penalties on home turf. In the EU, GDPR now dictates that firms must pay a fine of up to 4% of their annual global turnover if a data protection authority rules that they mishandled user information.

"We will see some of the principles that the Europeans have adopted but we'll see them implemented in a way that's more consistent with American political and legal traditions," he explained.

"But I think the tech sector itself ironically will find itself more supportive of rules that are similar to what we have in Europe. It would be so expensive and difficult to operate if we have to create a technical architecture that's fundamentally different from one we've created already."

This was supported by a senior source in the privacy world, who told BI that GDPR seemed radical in the US a year ago, but could now form the backbone of US legislation. The source added that privacy had been one of the major themes of Davos 2019.

Smith said: "In a world where we're all working with data I think we all have opportunities to ask how we can manage it better and in a way that sustains the public's confidence."

The next tech divide could be between those who have access to artificial intelligence and those who do not, Salesforce CEO Marc Benioff warned on Wednesday.

Artificial intelligence is becoming a "new human right," and everybody will need access to it, Benioff said in a speech at the World Economic Forum in Davos, Switzerland.

"Today, only a few countries and only a few companies have the very best artificial intelligence in the world," Benioff said. "Those who have the artificial intelligence will be smarter, will be healthier, will be richer, and of course, you've seen their warfare will be significantly more advanced."

By contrast, those who do not have access to AI will be "weaker and poorer, less educated and sicker," he said.

"We must ask ourselves, is this the kind of world we want to live in?" Benioff said. "This can be seen right where I live in San Francisco, where we truly have a crisis of inequality."

Debate has already emerged about the ethics of AI and the possibility that it can be used for harm, for example, in warfare. To tackle these questions and to guide its use of the technology, Salesforce in December hired its first chief ethical and humane use officer.

Read this: Salesforce is hiring its first chief ethical and humane use officer to make sure its artificial intelligence isn't used for evil

"AI is technology like none of us have ever seen, and none of us can truly say where it's going," Benioff said. "But we do know this: Technology is never good or bad. It's what we do with the technology that matters."

AI isn't the only divide the tech industry is grappling with. Companies in the industry are facing a crisis of trust in the wake of privacy breaches and the mishandling of data, Benioff said.

Benioff has becoming increasingly vocal about economic and other inequalities. Last fall, he was a major proponent of a measure in San Francisco that would tax large companies to help the city deal with its growing homelessness problem.

Some other tech figures followed his lead in making pledges to benefit the homeless, while other CEO's disagreed with the approach.

BuzzFeed is planning to cut 15% of its staff, according to a Wall Street Journal report on Wednesday.

The decision could cost 250 employees their jobs with the digital publisher, people familiar with the matter told The Journal.

Sources also said BuzzFeed's choice to trim its workforce could be an effort by the company to get closer to profitability ahead of a potential merger with another digital media outlet.

This story is still developing.

While many of us were making resolutions to bring in the New Year, Instagram influencer Caroline Calloway was paving the way for her redemption tour.

The 27-year-old Calloway has headlined the most recent news cycle after stories about her $165 "creativity workshop" gained traction in December. A viral Twitter thread accused Calloway of running a scam, and the Instagram star was forced to cancel her tour and refund attendees after she received heavy criticism.

Details emerged that Calloway, who has more than 800,000 Instagram followers, was wholly unprepared for the tour and even made people bring their own food to events. Supporters say Calloway was simply "in over her head," but that hasn't stopped her actions from drawing comparisons to the people behind the infamous Fyre Festival.

But last week, Calloway changed her mind: Along with the rest of the world — or at least Calloway's 800,000 followers — I found out that she had decided to un-cancel her creativity tour. The first rescheduled workshop would take place in New York this past Saturday — three days out from her announcement.

So I went.

Calloway's 'personal brand'

By many, Calloway is considered one of the first iterations of what we refer to as "influencers," the people behind those established social media brands who are able to make ripples across the Internet, just by posting content for their thousands of devoted followers.

Calloway first started to develop a following back in 2013, when she began documenting her picturesque life as an American expat studying at the "real-life Hogwarts" — also known as the University of Cambridge. Her posts were unique for their accompanying long, flowery captions describing romantic relationships and emotional break ups with Josh, then Oscar, then Conrad.

By spring 2015 she had amassed 300,000 followers, the Daily Mail reported.

"She is famous for something that didn't really exist until a few years ago: a personal brand," Man Repeller wrote about Calloway in June 2018. "Posting intimate personal details on social media is now commonplace, but when Caroline first started sharing stories about her life, her friends and her romantic relationships, it was different. Unique. A bit scandalous, even."

Her long captions read like excerpts from a young adult novel, and publishers agreed: Calloway nabbed a $500,000 deal with Flatiron Books in 2015 to write a memoir called "And We Were Like," based off the life she detailed on Instagram.

But that deal shuttered sometime later (her last post about the book is from April 2016). Calloway realized that the "the boy-obsessed version" of herself she painted on Instagram wasn't the one she felt comfortable conveying, she said during her workshop.

She backed out, and is still responsible for paying back the $165,000 advance she got for her book deal. (She joked during the workshop that she now takes Uber Pool instead of UberX because, "Hello, debt!")

Calloway's personal posts didn't end after she backed out of the book deal. But she leveled with her massive Instagram following in a post in November 2018, where she revealed she was feeling "broken and scared and still worthy of love," and archived two years worth of Instagram posts. She also shared on Facebook that she had struggled with an Adderall addiction during college.

The person that emerged, Calloway says, was her more authentic and true self. Instead of posts, Calloway informs followers about her day-to-day life through long Instagram Stories.

Her Stories contain lengthy blocks of text you might have to screenshot just to be able to read them in their entirety. She's used the Snapchat-like feature since it debuted in 2016 in a way much like she used her image captions: as a personal journal to share with the masses.

'Are you here for Caroline Calloway?'

So on Saturday morning, I found myself approaching a nondescript warehouse in Brooklyn's Bushwick neighborhood, double-checking Google Maps to ensure I had plugged in the correct address. There were no markings and no numbers on the building's exterior.

After messaging Calloway, I had secured myself an invitation to her un-canceled workshop. In a 1,500-word email I and other interested attendees received, Calloway shared the detailed itinerary for the five-hour workshop. She also shared what would change this time around: there would be no flower crowns, but there would be catered food.

"So can I guarantee you'll like this workshop? No," Calloway wrote in the email. "But I think there is a 95% chance you will, especially since you felt moved to buy this ticket in the first place."

The warehouse where Calloway's creative workshop was held. Paige Leskin

When I arrived at the warehouse, I had a fleeting thought that this was the scam itself: that all these people would show up to a building that didn't exist to attend a workshop that wasn't actually happening.

But as I approached the building, walking toward me with the same confused look were two 20-something women with blown-out hair, expensive-looking boots, and long, designer coats.

"Are you here for Caroline Calloway?" one of them asked me. The two women were lost, and had banded together to find the workshop.

I was definitely in the right place.

We were ushered into the warehouse by one of Calloway's assistants, and followed her up four flights of dark stairs. (Later, I learned that Calloway's assistants are two sisters in college, overwhelmed and overworked by how much time they've had to devote to helping Calloway put her un-canceled workshop together at the last minute.)

We followed the assistant into a gorgeous loft apartment filled with knick-knacks and plants, tailor-made for an Instagram photoshoot.

Calloway, left, speaks to workshop attendees.Kathleen O'Neill

As we entered, we were told to explore the space and locate our "personalized notebooks" before sitting down to talk with other attendees and grab coffee. The cover of my notebook was adorned with cheap, sticky letters spelling out my name, and an envelope inside contained scrapbook-ready stickers you could use to decorate your notebook.

The first hour of the five-hour workshop was devoted to "new student orientation," which Calloway said she wouldn't attend because she didn't want to "steal focus."

During the orientation, I chatted with some of the women seated around the room. Besides the three journalists in the room, there was a woman in Yale University's nursing program who had traveled down from Connecticut for the day to attend. There was an aspiring actor studying psychology in the city, and two woman who had flown in from Seattle for the workshop and were making a weekend of it in New York. Another woman said she had just quit her job, and had bought a one-way ticket to Los Angeles for next week.

Four of the women there were given scholarships to cover the costs of the $165 workshop. Some of these scholarships were able to be offered by charging reporters to "cover a sensationalized news-storm of their own making," Calloway's assistant told me in a text.

Attendees were treated to coffee and oat milk from Rude Health, a London-based company. Rude Health said that although it doesn't have a "paid partnership" with Calloway, the company "sent Caroline off with lots of Rude Health stocks for her American adventure." Paige Leskin

After an hour of mingling, Calloway arrived, albeit a little late. She showed off her white t-shirt that read "SCAMMER," and pointed out the Fyre Festival banner she proudly says she made herself, without using stencils.

"It was not a part of my dream to be compared to a literal Caribbean island where people almost died," Calloway said.

Calloway made her way around the room, stopping for long introductions and intimate conversations with some of the groups. Her prep for the event was evident: she came in knowing each person's last name, and was ready with remarks about any mutual friends or interests. She smiled widely and cracked cheeky jokes, letting attendees in on secrets like they were her friends. She was bubbly and easy-to-like, if a bit calculated.

During the workshop, Calloway said her class would cover topics like resiliency, creativity, heartbreak, and authenticity. In reality, this translated to long narratives about her life — many of which her fans already knew, since she had told them in past Instagram captions.

She also shared philosophical one-liners like, "You cannot read that doubt like tea leaves," and, "Sometimes closure is picking up a pretty red leaf and putting it on a bench and walking away." At one point, she compared sex to Thai food: "If you go to a restaurant and order Thai food and don't like it, you shouldn't keep eating Thai food."

Calloway also talked about her past. She talked about her addiction to Adderall in school, which she hadn't revealed much about online. She said the book deal "suffocated" her. And she insisted she's not trying to scam anyone, despite what the media says about her.

"People make a lot of assumptions of young, fit, white girls on Instagram," Calloway says. "You know what, I don't even read the news. I haven't read about what people think I am."

'I just totally connected with you on another level'

Many of the women in attendance told Calloway they had been following her since 2015. Several referenced a Total Sorority Move article from March 2015 that called Calloway's Instagram a fairy tale — "if a fairy tale consisted of drinking wine, flirting with boys, and studying in Europe."

These women said they felt a connection with the raw emotion she described in her captions. (Calloway maintains she was one of the first people to post a "crying selfie" to the Internet.) Several of the woman said they were drawn in by Calloway's authenticity, and that she seemed to just "get it."

"I just totally connected with you on another level," one attendee told Calloway.

That connection does not seem lost on Calloway. She described her life as a "journey" she and her fans had shared, that her highs and lows were something everyone in the room experienced. She painted a picture of "us versus the world" — Calloway and her followers on one side, the "haters" on the other.

"We've been through some crazy f-----g s--t together," she said at one point during the workshop. "You guys are in your own category of people I'll never forget."

Her fans have stuck behind her loyally, even as stories have referred to her as the creator of "the next Fyre Festival," or have — as Calloway calls it — "hate-followed" her on Twitter. Multiple attendees told me they had enjoyed the workshop, and said it was worth the $165 fee.

"I think she is someone who is learning and growing like the rest of us," one attendee told me in an Instagram message after the workshop. "I didn't expect too much from it after the rescheduling, but she is very relatable and kind. She went out of her way to remember everyone's first and last names as well as their letters they've written."

"Caroline did a brave thing. She wanted to offer her time, her heart, and her experiences to a community that she has quite literally grown up with," one of Calloway's assistants told me after the event. "Even though there were details of the tour and workshop that weren't perfect, I think she did a good job of making something beautiful for the people that came."

Later, after attendees took their solo portraits with Calloway, each person was given a "care package" to remember the workshop by: another notebook, a mason jar, and small colorful drawstring pouches holding a small candle, a "crystal" rock Calloway swears by, a bunch of flower seeds, a matchbook with a "Calloway House" crest, a face mask, and a stick of incense.

Journalists attending the event were also given an emergency thermal blanket — Calloway said she wanted the blankets to be a tongue-in-cheek nod to Fyre Fest.

The "care package" that workshop attendees were given at the end of the day, which included a notebook, incense, and a mason jar.Kathleen O'Neill

At one point during the workshop, Calloway interrupted her lesson to take a video for her Instagram Story.

In the video, she pans around the room of eager workshop attendants sitting in front of her. "What do you guys think?" she asks them.

"10 out of 10!" someone shouts.

She then turns the camera back on herself and deadpans: "Total f-----g scam, right?"

Tesla laid off 7% of its employees on Friday, seven months after cutting 9% of its workforce.

Tesla had expanded its workforce by 30% in 2018 as it ramped up production of its Model 3 sedan, CEO Elon Musk said in an email to employees on Friday. He suggested this round of layoffs was necessary for Tesla to become consistently profitable while introducing lower-priced vehicles, such as the long-awaited $35,000 version of the Model 3.

But Musk had framed the 2018 layoffs as a decision Tesla would not have to repeat.

"We are making this hard decision now so that we never have to do this again," he told employees at the time.

Read more: Everyone who's telling you that Tesla is influencing the rest of the auto industry is completely wrong

Five former Tesla employees who were let go on Friday spoke with Business Insider about their time at the time company and how they were told they were being laid off.

Each former employee requested anonymity because of fear of retribution from Tesla.

Were you affected by the Tesla layoffs? Have a story to share? Contact Business Insider's Linette Lopez at This email address is being protected from spambots. You need JavaScript enabled to view it. and Mark Matousek at This email address is being protected from spambots. You need JavaScript enabled to view it..

'Like sheep to the slaughter'

One former employee, who worked in Tesla's delivery division in California, woke up on Friday to Musk's email. He said he tried not to think about it but found that difficult once he arrived at his office, where his coworkers were discussing the email. He noticed one of his managers was being quiet and "kind of standoffish" as he arranged chairs in the corner of the office farthest from the employees.

The manager told the delivery employee and about 15 of his colleagues to meet him by the chairs. There, the manager read from a script, informing them they were being let go. The script sounded as if it had been written by Tesla's human-resources department, the delivery employee said.

"I sat there in silence like a sheep to the slaughter," he said.

One colleague asked if he could speak with the manager one-on-one after the speech, but the manager declined his request.

The manager told the employees their positions had become redundant, but not everyone on the delivery employee's team was let go, and the manager didn't provide specific details about why each employee was being laid off. The employees weren't given details about their severance packages, only a blank piece of paper on which to write their names and personal email addresses. The delivery employee then said his goodbyes, packed up his personal items, and left.

"I just want to know why we were the people who were picked," he said.

The delivery employee said being laid off wasn't devastating since he had planned to leave the company later in the year, but the abruptness of the announcement bothered him.

"It's a very crappy way to treat your employees who have worked their butts off," he said. "Now, I have no idea what to do."

Overall, the delivery employee said he enjoyed his time at Tesla, especially the camaraderie with his coworkers, but he didn't feel that same bond with his managers and sometimes felt pressure to work long hours with no advance notice.

'I was in shock'

An engineering employee who worked at the Tesla Gigafactory in Nevada said she was surprised by the layoffs, citing Musk's assertion that the 2018 layoffs would be Tesla's last.

About an hour after she arrived at work on Friday, management-level employees began pulling some of her colleagues into meeting rooms. An hour later, she was called into one of the rooms, told she was being let go, and asked to confirm her contact information. She was then accompanied to her desk, where her laptop was taken. She was directed to leave immediately.

The process took about five minutes, she said, and she was not given a reason for her termination.

"I was in shock," she said.

She was told she would receive two months' pay as severance, but hadn't seen the exact details of her severance as of Tuesday afternoon.

The engineer said she had liked her job at Tesla, particularly the ability to work with and learn from a variety of departments, but she wishes she had been given more information about why she was let go.

'It was actually a lot easier than I thought it was going to be'

A third former employee, who worked in Tesla's energy division in California, said that while she loved working at Tesla and was sad to see her time at the company end, she found it easier to be let go than to watch her colleagues be laid off and deal with the aftermath.

"Having seen everybody go through this before, being one of the people that remains, it is a lot harder than being one of the people that gets to go," she said.

After waking up on Friday morning, she read Musk's email and saw her manager had sent her a Google Calendar invite. Her meeting with her manager was brief, and she didn't ask why she had been let go since a conversation she had with another manager before the meeting suggested her manager hadn't been able to choose which members of his team would be laid off. She found the experience to be relatively painless because her team works remotely and wasn't in the office that morning.

"It was actually a lot easier than I thought it was going to be," she said.

She described Tesla as a demanding but rewarding work environment, and she said she felt pride in Tesla's mission and a close bond with her coworkers.

She doesn't mind having a few months of paid time off, she said, and the flood of messages she has received about professional opportunities has removed any anxiety she might have about her next step.

"People seem pretty eager to help me find a new job," she said.

It was another 'big purge'

A delivery-experience specialist who was let go told Business Insider that this round of layoffs was just another one of Tesla's big "purges." They happened every six months or so at the company, she said. Employees would trickle out monthly too. Then every three months or so, there would be a jolt of new hires.

The logic of who was let go, the woman said, "was random," and the turnaround was the fastest she'd ever seen at any company.

"The managers who were at my store a year ago are not the same managers today. My store operates as a sales, service, and delivery hub, and each role of management has been changed from January 2018 to January 2019. That's just management," she told Business Insider.

So, ultimately another purge was expected.

She described her seven months working at Tesla as exhausting and chaotic. Her job was to prepare the paperwork for customers coming in to buy their cars. She would then go to a document signing that could take five minutes or an hour, depending on the customer or how well the paperwork was done.

Then the customer would get a brief orientation of their car. According to her training, this was all supposed to take about 15 minutes, but it usually took much longer, and the result was a backlog of customers waiting. Sometimes she and her colleagues would have to get help from nearby vehicle centers.

Sometimes customers refused delivery of their cars because of damage, complicating matters even further, she said.

Going forward, she sees fewer cars being delivered because of a reduction in the tax credit offered to Tesla customers. She said she also worries about competition from new entrants in the electric-vehicle market, such as Porsche. But that doesn't mean she's lost faith in the company.

"I do have confidence that it's a great product. They have a good following still. They just need to hang on."

'So this is my walk of shame'

One supervisor at Tesla's Fremont, California, factory said his dismissal came as a total shock. He loved working at Tesla and found his colleagues to be incredibly supportive.

"I loved it, I loved what I did. It was fast-paced, exciting, very demanding. Challenging at best, but I loved it."

He was also good at it, he said. He said he received an outstanding review in December with an extremely generous performance reward. So, he said he had reason to believe he would make it through this round of layoffs untouched.

Plus, on the day of the layoffs, he was informed that his team would be saved, so he assumed he was included in that. But at the end of shift on Friday, his immediate manager asked him to gather his things and go with him to human resources.

"So this is my walk of shame?" he asked his manager.

He was asked to turn in his laptop and expect paperwork from the company on Saturday.

As for his 2018 performance bonus, he said he was disappointed. Because of Tesla's bonus structure, he will receive only a portion of his 2018 bonus.

A Tesla representative said the company offers cash options that vest quarterly over a four-year period.

"I worked hard, I got a great bonus, and they just took it away from me," the supervisor said.

The 2019 World Economic Forum kicked off in Davos, Switzerland, this week. As the world's business and political leaders descend upon the Alpine town, their armada of private jets fills the tarmac at nearby airports.

The Air Charter Service expects that roughly 1,500 private jets will shuttle attendees to and from the conference this week. According to the private-jet leasing company, that's up from the 1,300 private jets that serviced the WEF in 2018.

"The global interest in the event led us to analyze the private jet activity over the past five years of WEF," Andy Christie, private jet director at ACS, said in a statement. "Davos doesn't have its own airfield and, whilst we have several clients who fly into the town by helicopter, the four main airfields that private jet users attending the forum use are Zürich, Dübendorf, St. Gallen-Altenrhein, and St. Moritz."

Read more:The next CEO of Airbus explains the major hurdle that stands in the way of electric planes.

Oliver Cann, the head of strategic communications at the WEF, pushed back against the ACS's projections. Instead, Cann argued in a post on the organization's website that it expects private jet traffic to drop by 14% from 2018.

While the ACS's figures used data from four airports, the WEF projection is based only on traffic data from Zürich and St. Gallen-Altenrhein airports.

According to Christie, the most popular aircraft are Gulfstream GV and Bombardier Global Express planes. These are large, ultra-long-range private jets that typically cost more than $60 million per aircraft. With the ability to make discreet transoceanic flights at high speed, these aircraft are common among business leaders:

IBM's stock got a nice lift on Wednesday, rising nearly 9%, a day after the company announced its first annual revenue growth since 2011.

But analysts are divided over whether the software and services giant's 2018 report card is a sign of even better things to come, or a mere blip masking IBM's ongoing challenges.

While IBM's full year revenue growth increased 1% in 2018, revenue in the fourth quarter declined 3.5% year-over-year.

"It's looking like another mediocre IBM earnings release," Andrew Bartels, VP and principal analyst at Forrester Research, told Business Insider. "The report of strong cloud growth should be taken with a grain of salt. It's another quarter of IBM basically moving sideways."

Under the leadership of CEO Ginny Rometty, IBM has moved to transform its business, shedding older products and units and focusing on higher-growth areas such as cloud computing, AI and blockchain.

Although IBM's strength in the fourth quarter lay in consulting, its growth in Cognitive Solutions — the segment that includes its AI platforms, including IBM Watson, as well as certain database products — is what investors care most about.

This past quarter, the revenue of Cognitive Solutions was flat as compared to last year, but it did grow 2% at constant currency. This was the key growth analysts were looking for, and IBM was able to hit that target.

"We do think the software turn is going to be something that's repeatable," John Roy, lead analyst at UBS, told Business Insider. "They are going to see growth in software. That's pretty much a key element of the story for us. We'll see how sustainable that is."

While businesses like cloud computing and AI offer a lot of growth potential, they also put IBM in competition with fierce competitors.

IBM "barely poses a major challenge to today's tech giants, tough competitors in the new economy who have grabbed and continue to seize the major share of new growth areas that are so crucial for IBM's success," Investing.com senior analyst Haris Anwar said in a statement.

Read more:IBM soars as high as 7% after blowing away Wall Street expectations and showing annual revenue growth for the first time since 2011

A big theme this year will be IBM's upcoming acquisition of Red Hat, which is expected to close in the second half of 2019. Analysts say a big question will be whether IBM can successfully sell Red Hat's services and if Red Hat can help IBM create a working hybrid cloud — one that processes information across both public cloud offerings and private, on-premise servers.

Right now, IBM lags behind Amazon, Microsoft and Google in its cloud offering, but Red Hat could help IBM in tapping into customers who want to move to the cloud but are not ready to go all in — hence, helping it forge a stronger hybrid cloud.

"My aim isn't to downplay Red Hat's strong position in the cloud computing arena, but rather my problem remains squarely with the IBM's leadership which failed to foresee the changing tech landscape quickly enough," Anwar said.

Anwar warns that investors should avoid IBM and look elsewhere, but Roy disagrees. Roy said he's seen plenty of negativity in IBM stock, but he believes it's overdone. Instead, analysts' expectations are more flat, and there's room for marginal growth.

"All in all, we think that '19 could be good, and obviously the Red Hat acquisition is something that's sitting out there in the second half," Roy said. "Between now and when that closes, it looks like the rest of the businesses are doing OK and improving."

For all the hoopla around Netflix and Hulu, there's a big untapped market for free, ad-supported video, and that's why Viacom bought Pluto TV, Wade Davis, CFO of Viacom, told Business Insider.

Viacom paid $340 million for Pluto TV, which is a leading ad-supported streaming service. Five-year-old Pluto TV claims 12 million monthly active users, up from 5 million in 2016, and two-thirds of them are on connected TVs. It licenses programming from 130 film and TV partners, including Viacom. Viacom sees Pluto TV as a way to make money off its archives by selling advanced TV advertising against its audience, half of which Pluto says is 18-34.

Even with the buzz around Netflix and Amazon originals, people are hitting their limit for paid TV and often just want someone else to make the viewing decision for them, Davis said. That's what makes it a "perfect complement to Netflix," he said.

Read more: Big brands like Verizon and Toyota are backing Amazon's Freedive as the e-commerce giant pushes deeper into OTT advertising

"People are developing a little bit of subscription fatigue," Davis said. "A lot of what people historically want to do is they have 30, 40 minutes to kill and they just want to be entertained. Pluto, because it's this very familiar, friendly interface, is the perfect complement. There is a segment of the market that doesn't want to pay for anything, and we can be their video and entertainment choice."

Pluto TV took in under $100 million in revenue in 2018 as it just began selling its ad inventory. Still, Davis thinks the business has the potential to be much more as it ramps up its sales efforts with the help of Viacom's 600-person sales force and takes aim at YouTube and other ad-supported video services.

"We think it's more than a billion-dollar business," Davis said.

Viacom also plans to roll Pluto TV out globally, starting with Spanish-language content in the US and Latin America.

The free streaming space is getting more crowded with other services coming from Amazon, Sinclair, and Comcast. But Viacom doesn't seem bothered by the competition as Pluto already is a leader in users and ad inventory that's brand-safe. Davis said he even sees Viacom working with AT&T on mobile distribution for Pluto.

Some have sniffed at Pluto TV for its content, which boasts channels with names like FunnyAF, Monstercat TV, and Horror 24/7.

"It's unlikely you will ever find 'Game of Thrones' on Pluto TV," said Bernard Gershon of GershonMedia. "Pluto, you're not competing with Disney, Amazon, Netflix; you're competing with the Tubis, Fubos — ad supported, lower-than-premium-quality content."

"The content is a bunch of movies like 'Legally Blonde' and 'Caddyshack' that were hits in the 80s or sitcoms that aren't 'Friends,'" said Alan Wolk, cofounder and lead analyst at TVREV. "Pluto's magic is, it's a linear stream and you can also watch stuff on demand. While they are ad-supported, it's still half of what you get on linear TV."

Davis said "there's always room for improvement" in Pluto TV's offerings. "Adding our entire library is going to be significant." But, he said, as a free service, Pluto doesn't charge a subscription fee that it needs to justify by spending Netflix-like billions on original shows and movies.

"We're not going to enter the fray and spend billions of additional dollars a year to create original premium content," Davis said.

It makes sense for Pluto TV to take a different tack from the subscription services, said Peter Csathy, founder of media advisory firm Creatv Media.

"Pluto's the Rodney Dangerfield of the AVOD space," Csathy said. "It is kind of a potpourri of content, but Viacom hopes to bring it new life with some of its programming. Consumers have a breaking point. [Viacom] feels can compete and still be a winner in a sea of SVODs. If they tried to be another Netflix, it'd be much more challenging."

The airline's average fleet age is about 4 years old.

Source: Air Fleets

When we talk about algorithms and automation, we can't assume that handing responsibilities over to a machine will eliminate human biases. Artificial intelligence, after all, is constructed and taught by humans.

MIT Media Lab researcher and Algorithmic Justice League founder Joy Buolamwini has made it her mission not only to raise awareness of bias in facial recognition software, but also to compel companies around the world to make their software more accurate and to use its capabilities ethically.

Buolamwini coined the term "the coded gaze." She says the coded gaze is a "reflection of the priorities, the preferences, and also sometimes the prejudices of those who have the power to shape technology," she told an audience at the World Economic Forum's 2019 annual meeting in Davos, Switzerland.

There are real stakes here. As she noted in a viral TED Talk and a New York Times editorial, it's one thing to have Facebook confuse people when analyzing a photo, but another when law enforcement or a potential employer is utilizing such software.

Buolamwini has Ghanian heritage, and as a graduate student, she found that facial recognition software she was researching could detect the faces of light-skinned classmates but not her own, darker skin. She showed the Davos crowd a shot from her TED talk, in which she demonstrated the way that same software recognized her face when she wore a plain white mask over it.

Different companies use different software for facial recognition, but the AI involved learns through images fed to it during its development. If it's shown primarily white, male faces, it will become an expert at identifying white, male faces and can pick up subtle details; meanwhile, it will struggle to identify or differentiate faces with different skin hues, and will have a more difficult time assessing female features.

Buolamwini and MIT colleagues embarked on a thorough assessment of leading facial recognition programs — those from Microsoft, Face++, and IBM — to see how they fared. There were differences in exactly how accurate each program was, but all three generally had the same order of best to worst accuracy: lighter male, lighter female, darker male, darker female (Face++ had a .1% increase in accuracy for darker male than lighter male).

Before making her research public, Buolamwini sent the results to each company. IBM responded the same day, she said, and said developers would address discrepancies. She assessed IBM's updated software last year and found a notable improvement. The accuracy for correctly assessing darker males went from 88.0% to 99.4%, for darker females from 65.3% to 83.5%, and for lighter females from 92.9% to 97.6%, and lighter males stayed the same at 97.0%.

"So for everybody who watched my TED Talk and said, 'Isn't the reason you weren't detected because of, you know, physics? Your skin reflectance, contrast, et cetera,' — the laws of physics did not change between Dec. 2017, when I did the study, and 2018, when they launched the new results," Buolamwini said. "What did change is they made it a priority."

Buolamwini then recognized that while software primed for lighter males presented its own problems, even perfect software was at the whim of its handlers.

She noted an Intercept investigation published last summer that found IBM had worked with the New York City Police Department for years on a surveillance project utilizing facial recognition, which was never made public while in use. (The NYPD told the Intercept it never used the software's ability to distinguish skin color, IBM declined to comment on the project, and the NYPD ended its partnership with IBM in 2016.)

She then pointed out how hiring software like HireVue, a company that makes a program that allows large employers like Unilever and Hilton Worldwide to rapidly assess many job applicants' video interviews, also utilize facial recognition software. Boulamwini argued that if used without sufficient care, AI hiring programs could reinforce existing traits in a role rather than correct for them. (HireVue told Business Insider that testing for bias with the intention of eliminating it is an integral aspect of the way its software is used.)

"So it's not just the question of having accurate systems," she said. "How these systems are used is also important." It's why she launched the Safe Face Pledge, with four commitments: show value for human life, dignity, and rights; address harmful bias; facilitate transparency; and embed commitments into business practices. It launched with three signatories, but has yet to land a major corporation.

Facial recognition software is here to stay, and has potential for useful applications like protecting users from identity theft. As Buolamwini's research suggests, it poses less of a threat to citizens' rights the more accurate it is, and as its use becomes more widespread, companies must take the ethics around its use very seriously.

You can see Buolamwini's full presentation at WEF's website »

In November, Amazon Web Services announced a new service, based on the popular open source Apache Kafka software — a move taken by many industry insiders as a slap in the face to Confluent, a startup that's synonymous with Kafka, given that its founders literally created it.

For its part, though Confluent didn't back down. It introduced a new license to prevent cloud providers — like Amazon — from taking its open source software and reselling it for a profit. The move has proved controversial, with skeptics wondering if Confluent is doing more harm to the open source community than good.

Still, all of this hasn't scared off investors, it seems, as Confluent announced on Wednesday a $125 million funding round led by legendary venture firm Sequoia Capital. The deal values the startup at $2.5 billion.

It's a promising sign that investors still see upside in open source startups like Confluent, in an age when some in the industry view cloud titans as a threat.

What is Confluent?

Confluent is known for its namesake event streaming software, based on the open source Kafka, which processes events like orders, sales and customer experiences in real time, meaning that systems can make faster decisions. Kafka was started when Confluent's founders were still working at LinkedIn, and left to start the company after seeing lots of enthusiasm around it from the open source community.

"With Confluent, companies can connect all of their applications and data sources, enabling them to react and engage their customers in a faster, more personalized and more efficient manner," Jay Kreps, co-founder and CEO at Confluent, said in a statement around the funding.

In November, Amazon Web Services announced it would start offering Kafka on its cloud. It's completely legal to sell open source software — by its nature, open source software is free for just about anybody to use, for any purpose. Cloud companies like Amazon and Alibaba have taken advantage of this fact to use open source software as the foundation of their paid services for customers.

In turn, some companies, Confluent included, have responded by changing their licensing to prevent those cloud providers from using their software for commercial purposes.

"We think this is a positive change and one that can help ensure small open source communities aren't acting as free and unsustainable R&D for tech giants that put sustaining resources only into their own differentiated proprietary offerings," Kreps wrote in a blog post at the time of the licensing change.

It's proven to be a controversial move that's split the community: The companies say that it's a necessary move to protect their businesses and make sure that the cloud providers play fairly, but skeptics worry that placing limitations on open source software undermines the community's core principles.

Read more:After Amazon's cloud encroaches on its turf, a startup is taking a stand: Open source can't be 'free and unsustainable R&D' for tech giants

Open source is big business

Still, despite the risk that the big cloud providers will eat their lunch, open source software companies like Confluent have thrived in recent times. Companies like Elastic and MongoDB that built their businesses around open source software have gone public in successful recent IPOs.

For its part, Confluent, which has been used by companies like AUDI and Capital One, has now raised $206 million in total, and says that in the past year, its subscription bookings grew by a factor of 3.5. Kreps says in a blog post that Confluent is still dedicated to Kafka, and that the rising interest in the software is only benefetting the company in the long run.

"Apache Kafka has a massive community of contributors of which we're just one part," Kreps wrote in a blog post about the funding. "And the larger idea of building around events is a topic with interest and participation from hundreds of thousands of developers. We're proud to be a part of this growing movement and to do our part to make it a reality."