“Social media platforms and specific engagement algorithms breathe new life and new color into age-old antisemitic beliefs and antisemitic biases.”
At the JFNA General Assembly, CyberWell CEO & Founder Tal-Or Cohen Montemayor highlighted progress in removing antisemitic content while emphasizing the need for greater transparency around recommendation algorithms, stronger platform enforcement, and digital literacy to address the real-world consequences of online antisemitism.
Read more: cyberwell.org/news-releases/…
A Holocaust survivor dies from injuries sustained in the Bondi Beach terrorist attack in December 2025. The response from some users online? Celebrating her death.
Following Eva Geller’s passing, CyberWell identified posts mocking her death, denying the Holocaust, and glorifying the attack.
CyberWell has flagged this content for review, working to ensure that the glorification of antisemitic violence does not go unchallenged.
May her memory be a blessing.
1
71
AI has the potential to transform content moderation, but can it recognize antisemitism hidden behind coded language, memes, and seemingly harmless symbols?
The Jerusalem Post highlights CyberWell’s latest recommendations to social media platforms on improving AI-powered content moderation by combining advanced technology with real-world expertise.
Read the full article: jpost.com/diaspora/antisemit…
CyberWell welcomes the @oversightboard's call for safer platform design, more accurate AI moderation, and meaningful accountability.
In a recent submission to the Board, CyberWell outlined practical solutions, including expert-informed AI evaluation, stronger multilingual moderation, and human review to improve the detection of online hate: cyberwell.org/reports/cyberw…
Read the Oversight Board's position paper: oversightboard.com/news/beyo…
3
51
Halloween is meant to be fun. Turning Jews into antisemitic caricatures is not.
Every Halloween, CyberWell sees the same disturbing trend: people dressing as Jews and portraying them as greedy, money-obsessed, and evil.
These costumes recycle centuries-old antisemitic stereotypes, normalizing the dehumanization and vilification of Jews under the guise of Halloween fun.
Antisemitism is not a costume.
2
81
Jews blamed for orchestrating protests in France. Jews blamed for suppressing protests in Albania.
CyberWell uncovered conflicting antisemitic conspiracy theories spreading on social media, including coded antisemitic language, calls to attack Jewish businesses, and demands for another Holocaust.
Different narratives. Same antisemitic scapegoating.
Read our latest findings: cyberwell.org/news-releases/…
51
Online antisemitism is far more than hateful content on social media. It fuels conspiracy theories, threatens Jewish communities, and undermines social cohesion, with consequences that extend well beyond the digital world.
CyberWell was pleased to welcome a delegation from Azerbaijan, Kazakhstan, and Kyrgyzstan including tech experts, media professionals, and government officials.
From AI-powered solutions to stronger platform accountability, our discussions focused on what it will take to make the digital world safer for Jewish communities and society as a whole.
3
65
“Antisemitism constantly evolves through coded language, symbols, and cultural references designed to evade automated detection. AI moderation systems must draw on contextual and subject-matter expertise to recognize the antisemitic message, not just the words," said CyberWell CEO & Founder Tal-Or Cohen Montemayor.
Read full news release and the submission here: cyberwell.org/news-releases/…
1
37
Antisemitic conspiracy theories spread within hours of the FlyDubai hijacking attempt, reaching hundreds of thousands of people online.
CyberWell identified the rapid spread of false-flag conspiracies seeking to turn victims into perpetrators and fuel hostility toward Jews and Israelis.
As our CEO & Founder Tal-Or Cohen Montemayor explains, these dangerous narratives underscore the urgent need for social media platforms to strengthen their policies and enforcement.
Thank you to @kinsellawarren and Canada's @nationalpost for shining a spotlight on this growing threat.
Read the full article: nationalpost.com/opinion/col…
2
53
A CyberWell team attended #TikkunTech in Brussels, a two-day hackathon bringing together experts and innovators to develop technology-driven solutions to combat online antisemitism and foster Jewish life.
One key takeaway from the event was that fostering Jewish life online requires more than content moderation.
Moderating and removing antisemitic content remains critical, but platforms also need to address the systems that determine how harmful content is recommended, amplified, and repeatedly surfaced to users.
Greater transparency around recommendation systems and stronger safeguards against the algorithmic amplification of online antisemitism should be treated as core Trust & Safety issues.
This is an issue we at CyberWell have been monitoring closely, particularly when it comes to the exposure and reach of antisemitic content online, with our most recent findings highlighted in our report on AI-generated antisemitism: cyberwell.org/reports/ai-gen…
1
63
CyberWell recommends:
• Applying hate speech and violence policies equally to AI-generated content
• Strengthening detection and enforcement across formats
• Addressing coded antisemitic narratives and disclaimer-based evasion
Full report: cyberwell.org/reports/ai-gen…
Three years after the October 7 massacre, we remember the 1,200 innocent people murdered on that horrific day.
Today, we continue to confront the digital aftermath of October 7, including the most successful hijacking of major social platforms by terrorists, who livestreamed their atrocities to the world, followed by a wave of violent Jew-hatred that spread rapidly across the digital space.
We remember the victims, honor the survivors, stand with the families whose lives were forever changed, and pay tribute to all those who rose to the occasion in response to such profound evil. And we recommit ourselves to confronting the hatred that accompanied the massacre, countering its continued spread online, and being part of the solution.
May the memory of those murdered be a blessing.
Three years after October 7, tech platforms still have a terrorist propaganda loophole.
Content that would be prohibited when posted by a terrorist organization can still reach mass audiences when substantially the same material is redistributed by media accounts.
In a new op-ed for @jns_org, CyberWell’s CEO & Founder Tal-Or Cohen Montemayor and Research Partnerships Lead Gal Kadosh argue that platforms need clearer standards to distinguish reporting from redistribution, and stronger safeguards against the amplification and reuse of terrorist-produced content.
After 9/11, broadcasters had to decide how much airtime to give Osama bin Laden. In 2026, social platforms have no airtime to run out of. That makes their responsibility greater, not smaller.
Read the full piece here:
jns.org/opinion/tal-or-cohen…
Calls for or justification of violence against Jews are more than twice as likely to appear in antisemitic AI-generated content than in ordinary user-generated antisemitic content.
Part 4 of 5 unpacking key findings from the report: cyberwell.org/reports/ai-gen…
1
1
46
“False Flag” Antisemitic Conspiracies Spread After Attack on FlyDubai Flight
CyberWell CEO & Founder Tal-Or Cohen Montemayor responds to the antisemitic conspiracy theories emerging in the aftermath of the attack:
“Today’s horrific attack aboard a FlyDubai flight from the UAE to Israel is an ugly reminder of how quickly online antisemitism can distort the facts of an unfolding event. After a co-pilot allegedly stabbed the second pilot and attempted to take control of the aircraft, Israeli passengers helped restrain him and keep the injured pilot alive before the plane diverted safely to Saudi Arabia.
As news of the incident spread, ‘false flag’ claims quickly emerged on social media, alleging that Israelis had hijacked the plane or staged the attack. These claims reflect a broader pattern of conspiratorial self-victimization, an antisemitic conspiracy that falsely accuses Jews or Israelis of orchestrating attacks against themselves to manipulate public opinion. It is a contemporary expression of Jew-hatred and a form of misinformation aimed at spreading hatred, distrust and disgust toward Jews.
Platforms and LLMs cannot rely on poisoned knowledge sources or automated systems that fail to recognize these rapidly evolving forms of antisemitic conspiracy. They need real-time expertise from organizations like CyberWell that can identify emerging narratives, understand the context behind them and help platforms respond effectively.
‘False flag’ conspiracy theories are particularly corrosive because they do more than deny an attack. They recast victims as perpetrators and can perpetuate violence by fueling hatred and encouraging further violence against Jews.”
News release: cyberwell.org/news-releases/…
What does AI-generated antisemitism actually look like?
Here are four examples identified in CyberWell’s new report, showing how generative AI is being used to repackage and amplify longstanding antisemitic tropes.
Part 3 of 5 highlighting what our research uncovered.
🔗 Read the full report: cyberwell.org/reports/ai-gen…
2
52
AI-generated antisemitism is no longer a future concern. It's a rapidly growing reality.
CyberWell analyzed 307 verified cases of AI-generated antisemitic content posted across major social media platforms between January 2025 and February 2026.
One finding stands out:
⚠️ 98.4% of all identified content appeared in the latter half of the period analyzed.
The rise of accessible AI tools is making it easier than ever to create and spread antisemitic content at scale, often in highly engaging video, audio, and image formats.
Part 2 of 5 examining key findings from the report.
🔗 Read the full report: cyberwell.org/reports/ai-gen…
2
55
AI-generated antisemitism is rapidly emerging as a new challenge across social media.
“Generative AI now allows bad actors to industrialize hate, producing high-impact content that can reach millions, with enforcement often coming only after it has already been widely amplified,” said CyberWell CEO & Founder Tal-Or Cohen Montemayor.
📊 CyberWell’s report analyzed 307 verified pieces of AI-generated antisemitic content that collectively reached more than 30 million views.
📈 And the trend is accelerating: an astounding 98.4% of the content identified appeared in the latter half of the period analyzed.
This first-of-its-kind research examines the intersection of generative AI abuse and social media, showing how AI-generated images, videos, and audio are enabling antisemitic content to be produced and amplified at scale across major platforms.
Part 1 of 5 exploring key findings from the report.
🔗 Read the full report: cyberwell.org/reports/ai-gen…
1
69
Twenty-five years after 9/11, antisemitic propaganda is still trying to rewrite the story of the attacks.
As CyberWell CEO & Founder Tal-Or Cohen Montemayor told the National Post:
“Twenty-five years after terrorists hijacked planes and turned them into weapons, there are still active efforts online to hijack the story of 9/11 itself. Antisemites and extremists use the attention around the 25th anniversary to push false conspiracies, shamefully attempting to turn a moment of remembrance into a vehicle for sowing conspiracy and discord.”
Thank you to @kinsellawarren and the @nationalpost for highlighting CyberWell’s findings on this dangerous form of online antisemitism.
Read the full column here: nationalpost.com/opinion/col…
Twenty-five years after 9/11, Jews are still being falsely blamed for the attacks.
Year after year, CyberWell has identified posts on X accusing Jews of orchestrating 9/11 and having advance knowledge of it, and using coded antisemitic imagery and language to spread these claims.
This reflects one of the most prevalent forms of online antisemitism in recent years: scapegoating Jews for major crises, conflicts, and global events through conspiracy theories.
As we marked the anniversary this week, CyberWell is continuing to monitor and expose this dangerous pattern.
Read our news release to learn more: cyberwell.org/news-releases/…
1
50



































