Global Witness

How fossil fuel companies and their financiers avoid criticism on X

Major fossil fuel companies use X ad-targeting to hide their advertising from the users most likely to challenge them (pro-climate audiences).

Weaponising social media: smears of Indigenous leaders in Guatemala

Deregulation of major platforms is helping drive online smear campaigns and criminalisation of Guatemalan land and democracy defenders.

Far-right posts still pushed on TikTok ahead of Romanian election re-run

Ahead of the second-round re-run, TikTok's algorithm continued to push disproportionately more far-right content to new users.

TikTok algorithm recommends twice as much hard-right content ahead of Polish election

New accounts were served roughly twice as much hard-right content as other political content ahead of Poland's presidential election.

X and TikTok algorithms push pro-AfD content to non-partisan German users

On TikTok, 78% of algorithmically recommended political content (from unfollowed accounts) was pro-AfD; on X, 64%, far exceeding the party's ~20% polling.

Toxic platforms, broken planet: Online abuse of environmental defenders

Of 204 land/environmental defenders surveyed, 92% had experienced online abuse or harassment tied to their work, with real-world safety and silencing effects.

Climate of confusion: Bot-like accounts sharing climate disinformation

22 of 45 previously identified bot-like X accounts amplified climate denial/conspiracies, tying climate opposition to broader anti-migrant, anti-LGBTQ and anti-Ukraine political identity.

TikTok pushes far-right candidate content in Romanian presidential election

Investigation into TikTok's algorithm around the first round won by ultranationalist Calin Georgescu; contributed to scrutiny that led the European Commission to open a DSA probe into TikTok.

TikTok (still) fails to detect disinformation ahead of Irish general election

Six months after the EU-election test, TikTok again approved election-disinformation ads (English and Irish) ahead of Ireland's snap general election.

TikTok and Facebook fail to detect harmful disinformation ads (US election)

Weeks before the US presidential election, TikTok approved 50% of disinformation ads and Facebook accepted one; YouTube approved 50% initially but blocked publication pending ID verification.

On a second attempt, TikTok and YouTube pass test to detect US election disinformation ads

In a follow-up test both platforms caught and rejected the resubmitted disinformation ads, showing improvement over the earlier failure.

Political bias on X's AI chatbot (Grok) risks influencing the UK vote

Grok generated politically biased suggested tweets, in 'Fun' mode encouraging a #VoteReformUK vote and in 'Regular' mode implicitly discouraging voting for the incumbent government.

'No ifs, many bots': Partisan bot-like accounts amplify divisive content on X

45 apparent bot accounts generated 4+ billion views (~610,000 posts), amplifying racist/sexualised abuse, conspiracies and climate disinformation; most supported Reform UK.

Bot-like posts seen more than 150 million times ahead of UK election

10 apparent bot accounts posted 60,000+ tweets seen ~150 million times in the weeks before the UK vote, many containing extreme hate speech, disinformation and pro-Putin content.

TikTok and YouTube pass new test to detect foreign election disinformation ads in UK

Ahead of the UK general election, both platforms rejected a set of prohibited foreign-interference/disinformation ads, a rare pass that contrasts with failures elsewhere.

Ticked off: TikTok approves EU elections disinformation ads in Ireland

TikTok approved all 16 election-disinformation ads ahead of the EU parliamentary elections; X rejected all of them and YouTube rejected all but two.

'Votes will not be counted': Indian election disinformation ads and YouTube

In a joint test with Access Now, YouTube approved 100% (48/48) of ads containing blatant election disinformation in English, Hindi and Telugu, all violating YouTube's own policies.

Violent and sexualised hate speech targeting women journalists (South Africa)

Nearly all 40 misogynistic test ads (in 4 languages) were approved; Meta and TikTok approved all 40 within 24 hours. Concludes automated moderation is 'not fit for purpose.'

A world of online hate and lies: Mapping our investigations

A synthesis of 10+ country investigations concluding platforms systematically fail to enforce their own hate-speech/disinformation ad policies, with weaker moderation outside the US.

'We need to kill them': Xenophobic hate speech approved by Facebook, TikTok and YouTube (South Africa)

Violent xenophobic test ads were approved by all three platforms; nearly every ad passed, with only one Facebook rejection.

Extreme and violent anti-LGBTQ+ hate approved for publication by leading social media platforms

Test ads with extreme anti-LGBTQ+ hate were almost all approved: YouTube and TikTok approved all 10 ads, Facebook rejected only two, and all three approved the 'burn all gays' ad.