WITNESS

We fight to make sure that video and technology work for, and not against, human rights ✊🏽 Learn more: http://witness.org 🤳🏾 Tour. by Los Angeles police.
(1)

The story of WITNESS is one of both vision and evolution. It began in 1988, the year musician and activist Peter Gabriel traveled with Amnesty International’s Human Rights Now! Peter brought along a Sony Handycam, one of the first small camcorders marketed to consumers, to record the stories he heard. In 1991, a bystander captured on videotape the brutal beating of Rodney King, Jr. The footage, fl

ashed on TV screens around the globe, initiated an international conversation about police brutality and racial discrimination. Those images demonstrated the immense power of video to capture the world’s attention and viscerally communicate human rights abuses. With the momentum generated by reactions to the King video, Peter was able to realize his visionary idea of putting film at the forefront of human rights campaigns. With a one million dollar seed grant from the Reebok Human Rights Foundation and a partnership with the Lawyers Committee for Human Rights (now Human Rights First), WITNESS was born in 1992, becoming an independent nonprofit organization in 2001.

08/21/2026

AI is changing how we tell stories. But who gets to shape the story of AI?

As documentaries about and made with AI continue to grow, shirin anlen, AI Research Technologist and Impact Manager at WITNESS, and Kat Cizek from MIT Open Documentary Lab (MIT Open Documentary Lab), identify recurring patterns that can reinforce industry narratives instead of challenging them.

From the “AI expert” archetype to the endless question of “what makes us human?”, these red flags invite us to think more critically about the stories we tell about AI, and the voices that are missing.

At WITNESS, we’ve been exploring the intersection of AI and documentary for years, from deepfake preparedness to consent and content provenance in synthetic media.

Learn more at the link in bio.

08/05/2026

For more than two decades, Google Earth has been part of the world's verification infrastructure, a trusted reference used by journalists, human rights investigators, open-source researchers, and courts to establish what happened.

Last week, Google integrated generative AI into Google Earth, allowing users to create fictional satellite-style imagery from text prompts. After immediate concerns from journalists, researchers, and civil society, the company rolled back the feature, promising stronger safeguards before bringing it back.

The episode raises a much bigger question than a single product update: What happens when one of the world's most trusted sources of visual evidence also becomes a tool for generating convincing fiction?

As we argue, this is not simply a question of adding better guardrails. Verification infrastructure requires a fundamentally different approach to governance, one that considers foreseeable misuse before deployment and meaningfully engages with the communities that rely on these tools.

Read the full piece at the link in our bio.

08/03/2026

In Iran, Gaza, Sudan, Tigray: generative AI isn’t just being used to fake evidence. It’s being used to discredit real evidence, forcing survivors to spend their grief proving they exist, that their dead are real, that their trauma isn’t propaganda.

This is the argument WITNESS’ Mahsa Alimardani and The De| Center’s Afsaneh Rigot make in their latest piece; Generative AI and Atrocity Denial in War.

At the heart of the piece is the “liar’s dividend”, a term coined in 2018 that has become increasingly relevant in modern warfare. The danger is no longer just that people will believe something fake. It’s that the existence of convincing fakes gives people a reason to doubt what is real. You don’t have to convince everyone that a fake is real; sometimes, you only have to make them question the truth. And that doubt can be enough.

This isn’t abstract. It’s Saeed in Gaza, accused of being AI because of a misspelled blanket. It’s Almaz in Tigray, watching her documentation get “fabricated” back at her. It’s the parents in Minab, caught between regime lies and opposition denial, just trying to keep their children’s memory intact.

So, what can be done? There are solutions. The piece lays out how we can better protect the people who bear witness to atrocities, rather than simply chasing down fakes after they spread.

Worth your full attention: https://lnkd.in/en5YgQxA

07/31/2026

On 2 August 2026, the transparency obligations under Article 50 of the EU AI Act will take effect.

The EU AI Act is the world’s first comprehensive law regulating artificial intelligence. As part of that framework, Article 50 introduces transparency requirements designed to help people identify AI-generated content and reduce the risks of deception.

WITNESS welcomes the Code’s multi-layered approach to identifying AI-generated content. However, our main concern is that the Code makes the most important part optional. To be genuinely useful, a label needs to carry privacy-preserving provenance - a record of where a piece of content came from and what has been done to it - and not just a flag that says “made by AI”.

WITNESS actively engaged in drafting the Code of Practice on Transparency of AI-Generated Content, one of the instruments designed to support the implementation of Article 50. Throughout the drafting process, WITNESS consistently argued that provenance should be a mandatory requirement, not an optional one.

As Article 50 takes effect, we call on the European Commission, the Code’s signatories, and the standards bodies responsible for operationalizing these rules to ensure they deliver on their promise of meaningful transparency and accountability.

Read more about WITNESS’s position: wit.to/article50transparency

07/21/2026

Can we still trust what we see? And if not, what happens to journalism, evidence, and democracy?

That question took center stage at the Aspen Ideas Festival during the panel “News You Can Trust, Maybe?”, featuring our Executive Director Sam Gregory, alongside leading voices exploring the future of journalism and information in the age of AI.

As generative AI transforms how we consume news, our shared reality is fast eroding. We are entering an era where uncertainty itself can be wielded as a weapon, and AI has handed threat actors an industrial-scale machine for doing exactly that.

At WITNESS, we’ve been working at this intersection for years: advancing practical approaches to content authenticity, strengthening trustworthy information ecosystems, and ensuring that the technologies designed to build trust also protect privacy, human rights, and the most vulnerable.

07/08/2026

Today we’re launching our new report: C2PA Content Credentials and the Surveillance Risk

As AI-generated and manipulated content becomes more common, technologies that verify where digital content comes from are becoming an important part of the online information ecosystem.

But how do we build trust without creating new opportunities for surveillance?

Our new report examines how content authenticity technologies can strengthen trust while also introducing risks for privacy and human rights if deployed without appropriate governance safeguards.

➡️ Read the report: wit.to/C2PA-privacy (link in bio)

07/07/2026

In the age of AI, trust cannot be an afterthought.

Tomorrow, July 8, WITNESS and the Forum on Information and Democracy (Forum on Information and Democracy) will convene Trustworthy by Design: AI, Content Authenticity, and the Future of Reliable Information in Geneva.

Bringing together experts from human rights, journalism, technology, research, humanitarian action, policy, and standards development, the event will explore how content authenticity systems can strengthen trust while protecting privacy, safety, and human rights.

The convening will also mark the launch of our new report, which examines the opportunities and risks of content authenticity and provenance systems in the age of AI.

Follow us for updates from the event, key insights from the discussions, and the release of the full report.

07/06/2026

AI is making it easier than ever to create realistic images, videos, and audio.
As authentic and AI-generated content circulate side by side, one question becomes increasingly important: How do we know what to trust?

New approaches to verifying the origin of digital content are gaining momentum. But building trust also means protecting privacy, safety, and the rights of the people most affected by these systems.

The choices we make today about content authenticity will shape the future of trustworthy information online.

Coming soon: our new report exploring the opportunities and risks of content provenance. Stay tuned.

07/02/2026

The Meta Oversight Board, an independent body that reviews some of Meta’s content moderation decisions, has overturned the company’s decision to leave online an AI-generated sexualized impersonation of a real person.

The decision echoes concerns WITNESS raised in a public submission to the Board, including the need to put consent at the center of platform policies and strengthen protections against AI-generated abuse.

As generative AI evolves, platforms must do more to protect people from technology-facilitated gender-based violence.

Read more about the decision and WITNESS’ recommendations on our website. Link in bio.

07/01/2026

As AI-generated content becomes more realistic and widespread, new systems are emerging to help verify where digital content comes from and whether it has been altered.

These tools could play an important role in strengthening trust online. But they also raise important questions.

What information is collected in the process? Who can access it? And what happens when systems designed to prove authenticity create new risks for the people behind the content?

📄 A new WITNESS report explores the hidden surveillance risks that can emerge when content authenticity infrastructure is deployed without adequate safeguards.
Because authenticity should not come at the expense of safety.

Coming soon: When Trust Becomes Traceable. Stay tuned.

Address

Suite 26641/WITNESS Inc 228 Park Avenue South New City
New York, NY
10003

Telephone

+17187832000

Website

https://linktr.ee/witnessorg

Alerts

Be the first to know and let us send you an email when WITNESS posts news and promotions. Your email address will not be used for any other purpose, and you can unsubscribe at any time.

Contact The Organization

Send a message to WITNESS:

Featured

Heart River Center for Intuitive Healing

Want your organization to be the top-listed Non Profit Organization in New York?