Facebook whistleblower Frances Haugen gives evidence to MPs scrutinising online safety bill

Facebook whistleblower Frances Haugen today told MPs Facebook was ‘unquestionably’ making online hate worst due to its algorithm ‘prioritising extreme content’ and ‘subsiding’ damaging adverts.

Ms Haugen said Facebook was ‘very good at dancing with data’ to suggest it was successfully clamping down on toxic content but in reality its ‘negligence’ is pushing users towards extremism and undermining democracy.

The data scientist’s appearance coincided with her release of a fresh trove of documents which sensationally revealed CEO Mark Zuckerberg ‘personally intervened’ to allow US right wingers to ‘way what they wanted’.

They also revealed in detail how the tech firm has ignored internal complaints from staff for years to put profits first, ‘lie’ to investors and shield Mr Zuckerberg from public scrutiny.

The so-called ‘Facebook Papers’ comprise of internal research she secretly copied while working at the social media giant’s integrity unit – and were first unveiled at a Senate hearing earlier this month.

Today, Ms Haugen said hate being spread on social media had put the world at the ‘opening stages of a horrific novel’ which will see more political violence and ethnic cleansing if nothing is done.

Turning to Facebook-owned platform Instagram, she said she was ‘extremely worried’ about its impact on children’s mental health and behaviour, and argued it could tackle the issue but didn’t because ‘they want to get them hooked’.

Today, the whistleblower was addressing a parliamentary committee scrutinising the government’s Online Safety Bill, which would place a duty of care on social media companies to protect users – with the threat of substantial fines of up to 10% of their global revenue if they fail to do so.

Opening the session, she said: ‘I am extremely, extremely worried about the state of our societies. I am extremely concerned about engagement-based ranking, which prioritises extreme content.’

Ms Haugen said Facebook was reluctant to sacrifice even a ‘slither of profit’ to make the platform safer, and said the UK could be particularly vulnerable because it’s automated safety systems may be more effective with US English than British English.

‘I am deeply concerned about their underinvestment in non-English languages and how they mislead the public in how they are supporting them,’ she said.

‘UK English is sufficiently different that I would be unsurprised if the safety systems that they developed primarily for American English were actually underenforcing in the UK. Facebook should have to disclose those dialectical differences.’

She one of the effects of Facebook’s algorithm was to give hateful advertising greater traction, meaning it was ‘cheaper’ for companies and pressure groups to produce angry messages rather than positive ones.

Ms Haugen described this process as ‘subsiding hate’.

Responding to this afternoon’s session, Home Secretary Priti Patel said ‘tech companies have a moral duty to keep their users safe’ following a meeting with Facebook whistleblower Frances Haugen.

Ms Patel said it was a ‘constructive meeting’ on online safety.

Firing off a barrage of devastating allegations that will further trash Facebook’s already tattered reputation, Ms Haugen said – 

• Facebook’s algorithm prioritises hate speech by showing people content based on how much engagement it has received;
• ‘Anger and hate’ is the ‘best way to grow’ on the platform, and said bad actors were playing the algorithm by making their content more hateful;
• The world is ‘at the opening stages of a horrific novel’ due to extremism spreading via social media unless regulators act;
• Facebook is reluctant to sacrifice ‘even slithers of profit’ to prioritise online safety and was ‘unquestionably’ making online hate worse;
• Children’s relationship with platforms like Facebook is an ‘addicts’ narrative’, with youngsters saying social media sites make them unhappy but they are unable to stop using them;
• Said Facebook could tackle this problem but ‘they don’t because they know that young users are the future of the platform and the earlier they get them the sooner they get them hooked’; 
• Argued platform had demonstrated ‘negligence’ and ‘ignorance’, but resisted the term ‘malevolence’ as this ‘implies intent’.

Speaking to MPs today, Ms Haugen likened failures at Facebook to an oil spill.

‘I came forward now because now is the most critical time to act,’ she told the select committee. When we see something like an oil spill, that oil spill doesn’t make it harder for a society to regulate oil companies.

‘But right now the failures of Facebook are making it harder for us to regulate Facebook.’

The whistleblower said she had ‘no doubt’ that events like the storming of the US Capitol would happen in the future due to Facebook’s ranking system prioritising inflammatory content.

She said the problem could get worse due to the social media giant prioritising the creation on large Facebook group’s so people spend more time on the network.

‘Facebook has been trying to make people spend more time on Facebook, and the only way they can do that is by multiplying the content that already exists on the platform with things like groups and reshares,’ she said.

‘One group might produced hundreds of pieces of content a day, but only three get delivered. Only the ones most likely to spread will go out.’

Ms Haugen said Facebook groups were increasingly acting as ‘echo chambers’ that are pushing people towards more extreme beliefs.

‘You see a normalisation of hate and dehumanising others, and that’s what leads to violent incidents,’ she said.

The whistleblower argued regulation could benefit Facebook in the long run by making it a ‘more pleasant’ place to be.

She added that Twitter and Google were ‘far more transparent’ than Facebook, as she called for Mr Zuckerberg to hire 10,000 extra engineers to work on safety instead of 10,000 engineers to build its new ‘metaverse’ initiative.

Ms Haugen said that ‘anger and hate’ is the ‘best way to grow’ on Facebook, and said bad actors were playing the algorithm by making their content more hateful.

‘The current system is biased towards bad actors and those who push Facebook to the extremes.’

The whistleblower urged ministers to take into account the harm Facebook does to society as a whole rather than just individuals when considering new regulation.

‘Situations like [ethnic violence in] Ethiopia are just the opening chapters of a novel that is going to be horrific to read.

‘Facebook is closing the door on us being able to act. We have a slight window of time to regain people control over AI – we have to take advantage of this moment.’

Ms Haugen urged MPs to regulate paid-for advertisements on Facebook, because hateful ones were drawing in more users.

‘We are literally subsidising hate on these platforms,’ she said.

‘It is substantially cheaper to run an angry hateful divisive ad than it is to run a compassionate, empathetic ad.’

The whistleblower said Facebook was reluctant to sacrifice ‘even slithers of profit’ to prioritise online safety.

Ms Haugen said systems for reporting employee concerns at Facebook were a ‘huge weak spot’ at the company.

‘When I worked on counter espionage, I saw things where I was concerned about national security and I had no idea how to escalate those because I didn’t have faith in my chain of command at that point,’ she said.

‘We were told to accept under-resourcing.

‘I flagged repeatedly when I worked on civic integrity that I felt that critical teams were understaffed.

‘Right now there’s no incentives internally, that if you make noise saying we need more help, like, people will not get rallied around for help, because everyone is underwater.’

Ms Haugen told the parliamentary select committee that the social media giant was ‘unquestionably’ making online hate worse.

‘We didn’t invent hate, we didn’t invent ethnic violence. And that is not the question.

‘The question is what is Facebook doing to amplify or expand hate … or ethnic violence?’

Ms Haugen said she ‘sincerely doubted’ that it was possible for Instagram to be made safe for children and that the platform promoted an ‘addict’s narrative’.

‘Children don’t have as good self regulation as adults do, that’s why they’re not allowed to buy cigarettes,’ she said.

‘When kids describe their usage of Instagram, Facebook’s own research describes it as ‘an addict’s narrative’.

‘The kids say ‘this makes me unhappy, I don’t have the ability to control my usage of it, and I feel if I left it would make me ostracised’.’

She continued: ‘I am deeply worried that it may not be possible to make Instagram safe for a 14-year-old and I sincerely doubt that it is possible to make it safe for a 10-year-old.’

Ms Haugen said Facebook could estimate people’s ages with ‘a great deal of precision’ but did not act to stop under-age users.

‘Facebook could make a huge dent on this if they wanted to and they don’t because they know that young users are the future of the platform,’ she told a parliamentary select committee.

‘The earlier they get them, the more likely they’ll get them hooked.’

Facebook has previously outlined plans to set-up a so-called Instagram Kids. It argues that under-13s already use Instagram despite age barriers, and that the new app would be safer for them.

Ms Haugen first aired her bombshell revelations in front of the US Senate earlier this month, where she argued a federal regulator is needed to oversee digital giants like Facebook.

The draft Online Safety Bill proposes something similar by creating a regulator that would monitor Big Tech’s progress with removing harmful or illegal content from their platforms, such as terrorist material or child sex abuse images.

Ministers also want social media companies to stamp down on online abuse by anonymous trolls.

Damian Collins, Chair of the Joint Committee on the Draft Online Safety bill, called Ms Haugen’s appearance ‘quite a big moment’.

‘This is a moment, sort of like Cambridge Analytica, but possibly bigger in that I think it provides a real window into the soul of these companies,’ he said.

Mr Collins was referring to the 2018 debacle involving data-mining firm Cambridge Analytica, which gathered details on as many as 87 million Facebook users without their permission.

The committee has already heard from another Facebook whistleblower, Sophie Zhang, who raised the alarm after finding evidence of online political manipulation in countries such as Honduras and Azerbaijan before she was fired.

It comes as concerns were raised that details of the new legislation could be leaked to Facebook  by civil servants who ‘want to work for government for four years before getting job at tech giants’

The alarm was raised after an online harms issue known only to a few people at the Department for Digital, Culture, Media and Sport was raised by a senior executive at Facebook in a recent meeting.

Jobs taken by former senior civil servants in the private sector are meant to be scrutinised by the Advisory Committee on Business Appointments (Acoba). But its powers are weak and more junior appointments are not vetted.

A source lashed out at department mandarins, telling the Times: ‘The problem is that DCMS officials think it’s their job to work there for four years then get a job at Facebook.

‘They don’t get scrutinised by Acoba except at the most senior level.’

Several DCMS officials have gone on to work for Facebook recently, after working elsewhere before joining Facebook.

There is no suggestion they have solicited information from former Civil Service colleagues.

Facebook whistleblower docs reveal it ‘has known for YEARS’ that it fails to stop hate speech and is unpopular among youth but ‘lies to investors’: Apple threatened to remove app over human trafficking and staff failed to see Jan 6 riot coming 

By Jack Newman for MailOnline

A trove of documents from Facebook whistleblower Francis Haugen have revealed in detail how the tech firm has ignored internal complaints from staff for years to put profits first, ‘lie’ to investors and shield CEO Mark Zuckerberg from public scrutiny.

The documents were reported on in depth this morning as part of an agreement by a conglomerate of media organizations, as Haugen testified before the British Parliament about her concerns.

They are comprised of internal research that she chose to make public. They are now being referred to by the US media as the ‘Facebook Papers’.

They claim, among other things, that;

• Facebook staff have reported for years that they are concerned about the company’s failure to police hate speech;
• That Facebook executives knew it was becoming less popular among young people but shielded the numbers from investors;
• That staff failed to anticipate the disastrous January 6 Capitol riot despite monitoring a range of individual, right-wing accounts;
• On an internal messaging board that day, staff said: ‘We’ve been fueling this fire for a long time and we shouldn’t be surprised it’s now out of control’;
• Apple threatened to remove the app from the App Store over how it failed to police the trafficking of maids in the Philippines;
• Mark Zuckerberg’s public comments about the company are often at odds with internal messaging.

Some of the most damning comments were posted on January 6, the day of the Capitol riot, when staff told Zuckerberg and other executives on an internal messaging board that they blamed themselves for the violence.

‘One of the darkest days in the history of democracy and self-governance. History will not judge us kindly,’ said one worker while another said: ‘We’ve been fueling this fire for a long time and we shouldn’t be surprised it’s now out of control’.

One of her complaints is how the company had been warned by staff for years that it was not doing enough to police hate speech.

One of the problems is its AI tools do not have the capability to appropriate pick out hateful commentary, and there aren’t enough staff with the language skills to do it manually.

The failures to block hate speech in volatile regions such as Myanmar, the Middle East, Ethiopia and Vietnam could contribute to real-world violence.

In a review posted to Facebook’s internal message board last year regarding ways the company identifies abuses, one employee reported ‘significant gaps’ in certain at-risk countries.

Facebook spokesperson Mavis Jones said in a statement that the company has native speakers worldwide reviewing content in more than 70 languages, as well as experts in humanitarian and human rights issues.

She said these teams are working to stop abuse on Facebook’s platform in places where there is a heightened risk of conflict and violence.

‘We know these challenges are real and we are proud of the work we’ve done to date,’ Jones said.

Still, the cache of internal Facebook documents offers detailed snapshots of how employees in recent years have sounded alarms about problems with the company’s tools – both human and technological – aimed at rooting out or blocking speech that violated its own standards.

The material expands upon Reuters’ previous reporting on Myanmar and other countries where the world’s largest social network has failed repeatedly to protect users from problems on its own platform and has struggled to monitor content across languages.

Among the weaknesses cited were a lack of screening algorithms for languages used in some of the countries Facebook has deemed most ‘at-risk’ for potential real-world harm and violence stemming from abuses on its site.

The company designates countries ‘at-risk’ based on variables including unrest, ethnic violence, the number of users and existing laws, two former staffers told Reuters.

The system aims to steer resources to places where abuses on its site could have the most severe impact, the people said.

Facebook reviews and prioritizes these countries every six months in line with United Nations guidelines aimed at helping companies prevent and remedy human rights abuses in their business operations, spokesperson Jones said.

In 2018, United Nations experts investigating a brutal campaign of killings and expulsions against Myanmar’s Rohingya Muslim minority said Facebook was widely used to spread hate speech toward them.

That prompted the company to increase its staffing in vulnerable countries, a former employee told Reuters.

Facebook has said it should have done more to prevent the platform being used to incite offline violence in the country.

Ashraf Zeitoon, Facebook’s former head of policy for the Middle East and North Africa, who left in 2017, said the company’s approach to global growth has been ‘colonial,’ focused on monetization without safety measures.

More than 90 per cent of Facebook’s monthly active users are outside the United States or Canada.

Facebook has long touted the importance of its artificial-intelligence (AI) systems, in combination with human review, as a way of tackling objectionable and dangerous content on its platforms.

Machine-learning systems can detect such content with varying levels of accuracy.

But languages spoken outside the United States, Canada and Europe have been a stumbling block for Facebook’s automated content moderation, the documents provided to the government by Haugen show.

The company lacks AI systems to detect abusive posts in a number of languages used on its platform.

In 2020, for example, the company did not have screening algorithms known as ‘classifiers’ to find misinformation in Burmese, the language of Myanmar, or hate speech in the Ethiopian languages of Oromo or Amharic, a document showed.

These gaps can allow abusive posts to proliferate in the countries where Facebook itself has determined the risk of real-world harm is high.

Reuters this month found posts in Amharic, one of Ethiopia’s most common languages, referring to different ethnic groups as the enemy and issuing them death threats.

A nearly year-long conflict in the country between the Ethiopian government and rebel forces in the Tigray region has killed thousands of people and displaced more than 2 million.

Facebook spokesperson Jones said the company now has proactive detection technology to detect hate speech in Oromo and Amharic and has hired more people with ‘language, country and topic expertise,’ including people who have worked in Myanmar and Ethiopia.

In an undated document, which a person familiar with the disclosures said was from 2021, Facebook employees also shared examples of ‘fear-mongering, anti-Muslim narratives’ spread on the site in India, including calls to oust the large minority Muslim population there.

‘Our lack of Hindi and Bengali classifiers means much of this content is never flagged or actioned,’ the document said.

Internal posts and comments by employees this year also noted the lack of classifiers in the Urdu and Pashto languages to screen problematic content posted by users in Pakistan, Iran and Afghanistan.

Jones said Facebook added hate speech classifiers for Hindi in 2018 and Bengali in 2020, and classifiers for violence and incitement in Hindi and Bengali this year. She said Facebook also now has hate speech classifiers in Urdu but not Pashto.

Facebook’s human review of posts, which is crucial for nuanced problems like hate speech, also has gaps across key languages, the documents show.

An undated document laid out how its content moderation operation struggled with Arabic-language dialects of multiple ‘at-risk’ countries, leaving it constantly ‘playing catch up.’

The document acknowledged that, even within its Arabic-speaking reviewers, ‘Yemeni, Libyan, Saudi Arabian (really all Gulf nations) are either missing or have very low representation.’

Facebook’s Jones acknowledged that Arabic language content moderation ‘presents an enormous set of challenges.’ She said Facebook has made investments in staff over the last two years but recognizes ‘we still have more work to do.’

Three former Facebook employees who worked for the company´s Asia Pacific and Middle East and North Africa offices in the past five years told Reuters they believed content moderation in their regions had not been a priority for Facebook management.

These people said leadership did not understand the issues and did not devote enough staff and resources.

Facebook’s Jones said the California company cracks down on abuse by users outside the United States with the same intensity applied domestically.

The company said it uses AI proactively to identify hate speech in more than 50 languages.

Facebook said it bases its decisions on where to deploy AI on the size of the market and an assessment of the country’s risks. It declined to say in how many countries it did not have functioning hate speech classifiers.

Facebook also says it has 15,000 content moderators reviewing material from its global users. ‘Adding more language expertise has been a key focus for us,’ Jones said.

In the past two years, it has hired people who can review content in Amharic, Oromo, Tigrinya, Somali, and Burmese, the company said, and this year added moderators in 12 new languages, including Haitian Creole.

Facebook declined to say whether it requires a minimum number of content moderators for any language offered on the platform.

Facebook’s users are a powerful resource to identify content that violates the company’s standards.

The company has built a system for them to do so, but has acknowledged that the process can be time consuming and expensive for users in countries without reliable internet access.

The reporting tool also has had bugs, design flaws and accessibility issues for some languages, according to the documents and digital rights activists who spoke with Reuters.

Next Billion Network, a group of tech civic society groups working mostly across Asia, the Middle East and Africa, said in recent years it had repeatedly flagged problems with the reporting system to Facebook management.

Those included a technical defect that kept Facebook’s content review system from being able to see objectionable text accompanying videos and photos in some posts reported by users.

That issue prevented serious violations, such as death threats in the text of these posts, from being properly assessed, the group and a former Facebook employee told Reuters. They said the issue was fixed in 2020.

Facebook said it continues to work to improve its reporting systems and takes feedback seriously.

Language coverage remains a problem. A Facebook presentation from January, included in the documents, concluded ‘there is a huge gap in the Hate Speech reporting process in local languages’ for users in Afghanistan.

The recent pullout of U.S. troops there after two decades has ignited an internal power struggle in the country. So-called ‘community standards’ – the rules that govern what users can post – are also not available in Afghanistan’s main languages of Pashto and Dari, the author of the presentation said.

A Reuters review this month found that community standards weren’t available in about half the more than 110 languages that Facebook supports with features such as menus and prompts.

Facebook said it aims to have these rules available in 59 languages by the end of the year, and in another 20 languages by the end of 2022.