Hate speech targeting Jews now appears in dozens of languages, across thousands of platforms, in posts, comments, replies, livestreams, and short-form video. Most of it disappears within hours, and much of it is never flagged at all. Adam Milstein decided years ago that this was a problem only software could solve. In April 2026, he made the argument explicit, calling in a Jerusalem Post op-ed for “an AI-powered global Jewish response network” to be treated as a communal priority.
The Adam and Gila Milstein Family Foundation, which now supports more than 200 nonprofit organizations and has produced over 1,000 collaborations among them, has built a portion of its portfolio around technology-driven monitoring of online antisemitism. The foundation backs CyberWell, an organization that uses AI to identify antisemitic content on major platforms, classify it against accepted definitions, and produce data the platforms themselves can no longer credibly ignore.
The thesis is straightforward. Traditional monitoring relies on volunteer flaggers, manual review, and small research teams trying to keep up with platforms generating billions of posts per day. The math has never worked. CyberWell’s approach inverts it: train models on what antisemitism actually looks like, run them against the firehose, and produce reports that put hard numbers behind what the Jewish community has been describing for years. Numbers, in Milstein’s experience, move institutions where stories do not.
Milstein’s April 2026 op-ed laid out a six-part framework for diaspora resilience that placed AI-enabled crisis infrastructure on par with physical security and Jewish education. “Every Jewish institution, and every individual facing harassment, should have access to coordinated networks providing legal, communications, and security support in real time,” he wrote. “This must include AI-enabled tools to counter the growing battlefield of information warfare.”
The piece was less a plea than a directive.
This is a pattern visible across his giving. Milstein has long argued that the threats facing Jewish communities require infrastructure, not just outrage. Other parts of the portfolio, including Palestinian Media Watch, HonestReporting, and NGO Monitor, apply analogous discipline to legacy media, NGO networks, and Arabic-language broadcasting. His own Hager Pacific Properties career was built on quantitative discipline: due diligence, performance metrics, ongoing oversight rather than one-time capital deployment. Funding an AI-driven antisemitism tracker isn’t a leap from real estate. It’s the same playbook applied to a different sector.
There is also a strategic case. The most dangerous antisemitic networks online are not the ones that go viral. They are the ones operating in closed channels, in coded language, in communities that platform algorithms surface to vulnerable young men before any human moderator catches them. Identifying those networks at scale requires pattern recognition no human team can match. That is precisely what modern machine-learning systems do well.
Milstein has not been quiet about the limits of the approach. He has argued, including in his December 2024 Jerusalem Post column, that technology is necessary but not sufficient. Software flags hate. Coalitions, legal action, education, and political pressure are what convert that data into consequence. The Impact Forum, which he and Gila co-founded in 2017, has been one venue for assembling that broader response, and the foundation’s CyberWell support has been one input feeding into it.
At 73, Milstein is funding tools he will not personally use at scale and platforms whose interfaces will look unrecognizable in a decade. That is, in his telling, exactly the point. The infrastructure has to be built before the next escalation, not after. The April 2026 op-ed closed with the line he has been making in private for years: “The future of Jewish life will not be secured by those who make the best arguments. It will be secured by those who are prepared to defend it, build it, and lead it.”
