Twitter broadens its definition of harm to address content with deceptive language around COVID-19, including “denial of established scientific facts”

Site changes rules to ban content aimed at making people act against official advice — Twitter will remove tweets that run …

The Guardian 2020-03-20 Alex Hern

Context & Ripple Effects

Ten days into the pandemic's declared spread, Twitter is moving from case-by-case takedowns to a written rule: 'harm' now covers deceptive COVID-19 language, including denial of established scientific facts and content designed to push people off official public-health advice. The change gives the company a codified basis to act rather than improvise.

The arc that follows is visible in the related coverage: within weeks Twitter was applying the new rules against prominent accounts including Bolsonaro and Giuliani, by May it had built the label-and-warning-message system for disputed claims, and by December it extended removals to vaccine misinformation. The endpoint came in November 2022, when Twitter stopped enforcing the policy outright after suspending more than 11,000 accounts under it.

First-order effects

Accounts spreading deceptive COVID-19 content face removal of tweets under a now-explicit rule, and high-profile violators — as the Bolsonaro and Giuliani actions showed — are no longer exempt.

Second-order effects

Rival platforms face pressure to match a written standard rather than ad-hoc moderation, while health authorities gain a lever: official guidance becomes the reference point the platform enforces against.

Third-order effects

The full cycle — rule written in crisis, scaled through labels and removals, then abandoned with 11,000+ suspensions on the record — shows platform health policies are only as durable as the leadership behind them, setting up the later reversals such as the 2023 violent-speech policy rewrite.

The trend: Platform content governance is oscillating between crisis-driven rule expansion and later deregulation, with COVID-19 misinformation as the clearest test case of how reversible moderation regimes are.

Discussion

Twitter Twitter on x
An update on our continuity strategy during COVID-19
@twittersafety @twittersafety on x
Content that increases the chance that someone contracts or transmits the virus, including: - Denial of expert guidance - Encouragement to use fake or ineffective treatments, preventions, and diagnostic techniques - Misleading content purporting to be from experts or authorities
@twittersafety @twittersafety on x
Update: we're expanding our safety rules to include content that could place people at a higher risk of transmitting COVID-19. Now, we will require people to remove Tweets that include the following:
@jamiejbartlett Jamie Bartlett on x
Twitter has hugely expanded its definition of ‘harm’ and will remove misinformation about #COVID19, including false diagnostic advice, claims that incite panic, claims that incite behaviour to benefit a third party, denial of established science & more. https://blog.twitter.com/.…
@alexhern Hern on x
Amazingly, Twitter still doesn't let you report something as misinformation, despite now banning misinformation. https://www.theguardian.com/ ...
@rosenbergerlm Laura Rosenberger on x
Wow. Showing the art of the possible when social media platforms have the will. https://twitter.com/...
@saminalrights @saminalrights on x
Twitter will actively remove harmful fake news about #coronavirus that puts others at risk. Excellent. Imagine if they did the same for climate deniers and animal ag industry lies. https://www.theguardian.com/ ...
@tha_rami Rami Ismail on x
Twitter will delete harmful info about COVID-19, & has expanded the definition of harmful to include scientifically inaccurate claims. COVID-19 is showing us clearly that a lot of things to make the world better were always possible, just not profitable. https://blog.twitter.com/…
@humanists_uk Humanists UK on x
Twitter says it will be applying a new broader definition of harm to address content that ‘goes directly against guidance from authoritative sources of global and local public health information’. #CoronaVirusUpdate #coronavirus #COVID19 https://www.theguardian.com/ ...
@wilsonleungws @wilsonleungws on x
Twitter has updated its policies to ban “denial of established scientific facts about transmission [of #COVID19] during the incubation period”. Does this mean that the WHO will have to delete its tweet of 14 Jan 2020 stating “no clear evidence of human-to-human transmission”? htt…
@jeremylittau Jeremy Littau on x
These are good changes. Sociology plays a role in the spread of disease. https://twitter.com/...
@qasimrashid Qasim Rashid on x
You know, Twitter, most of us are old enough to remember [checks notes] 19 days ago when 45 called #COVIDー19 a Democratic hoax & denied expert guidance—and he's still online. Your policies need to have actual consequences. https://twitter.com/...
@backlon Dieter Bohn on x
They say Twitter isn't real life but very rich people don't get punished for breaking the rules here either so it seems pretty real life to me. https://twitter.com/...
@sokane1 Sean O'Kane on x
NEW: Twitter says that after “reviewing the overall context and conclusion” of this Elon Musk tweet, the company decided it doesn't break its rules around coronavirus misinformation. https://twitter.com/...
@cwarzel Charlie Warzel on x
Tracking disinformation is much easier when things are black & white & we have scientific certainty. Right now we're clearly dealing with a huge lack of data and information. As docs will tell you, we don't know enough about this virus. Zeynep's mask argument is a great example. …
@zeynep Zeynep Tufekci on x
How will this apply to discussing advice from authorities that we believe are not supported by evidence? Can we contradict Trump? I just wrote a NYT op-ed arguing for universal mask wearing and got tons of medical people contacting me privately to say thank you. Do I get banned? …
@stacyherbert Stacy Herbert on x
Would Twitter have removed Boris Johnson's probable misinformation about ‘herd immunity’ as an option at this stage or his claims that he continues shaking hands? https://twitter.com/...
@iainmacwhirter Iain Macwhirter on x
They're going to be very busy. https://twitter.com/...
@stacyherbert Stacy Herbert on x
From what I can tell, there is not yet much ‘established science’ with #covid19? I mean, even something as simple as the efficacy of medical masks. The US Surgeon General claims 'they don't work,' while studies and scientists say otherwise. 🤔 https://twitter.com/...
@sachabaroncohen Sacha Baron Cohen on x
With coronavirus, lies and conspiracies will literally kill us. History will judge social media companies by what they do now. Thank you @Twitter for putting facts and science ahead of misinformation and profit. https://blog.twitter.com/...
@miekeeoyang @miekeeoyang on x
Great, now apply them to @realDonaldTrump. https://twitter.com/...
@ggreenwald Glenn Greenwald on x
I'll be very curious to see if this applies to world leaders like President Bolsonaro and various science-denying fanatics in his movement who continue to minimize the dangers of coronavirus and though various ways encourage people to gather in dangerous situations: https://twitt…
@tayhatmaker Taylor Hatmaker on x
Under Twitter's aggressive new COVID-19 safety rules, you can't encourage people to go to bars or discourage them from social distancing in areas affected by the coronavirus https://techcrunch.com/...

Chronicles