How the Electronic Frontier Foundation thinks about AI

You know the ways AI is regularly talked about—how much can it really do? How much will it cost? Environment? Bubble? We get that. But the Electronic Frontier Foundation wants to have a different conversation about AI. EFF's background on AI is deep. In 2017, we launched a detailed project to [Measure the Progress of AI Research](https://web.archive.org/web/20240420163406/https://www.eff.org/ai/metrics), encouraging machine learning researchers to [give us feedback and contribute to the effort](https://web.archive.org/web/20240422233351/https://www.eff.org/ai/metrics#How-to-contribute-to-this-notebook). That project was archived for lack of bandwidth, staffing, and the complexity and time required. But just five years later and the "progress of AI" is a global concern/topic, and everyone, including EFF, is thinking about it. Here's how \*we\* think about it, from the perspective of protecting civil liberties AND innovation. What do you think, and what are we missing? This is our summary: >AI technologies are affecting our civil liberties as never before. Ensuring that AI serves people, not power, starts with cutting through the hype. AI technologies are not magic wands—they are general-purpose tools. If we want to regulate those technologies to reduce harms without shutting down benefits, we have to focus on who uses AI, what products they use, and how they use them. >Where we see potential benefits, like improving weather forecasting, facilitating medical research, identifying systemic bias, or fostering accessibility, we work to ensure those benefits can be realized. >Where we see potential harms, we consider the practical and legal tools we already have, like pressure campaigns, privacy lawsuits, and transparency measures. If we need new tools, we should create protections tailored to the actual problem – not just to the latest outrage. For example, if policymakers are worried about AI accelerating systemic privacy violations, they should enact real and comprehensive privacy legislation that covers all corporate surveillance and data use, and close the data broker loophole to limit government surveillance. >And to keep the window open for a better future, we fight for a competitive innovation environment. For example, if we want AI models that don’t replicate existing social and political biases, we need to make enough space for new players to build them, and avoid giving today’s giants the power to block future competitors from offering us a better tool or product. >In research labs, conference rooms, courtrooms, and legislatures, people are making decisions that will determine who AI serves and how. EFF works to ensure those decisions support freedom, justice and future innovation. We have subcategories, as well. For example: AI and Surveillance. >AI tools amplify the threat of mass surveillance. By dramatically reducing the time and labor required to process massive amounts of personal data, AI increases the ability of governments and corporations to collect and act on invasive surveillance. Face recognition in all of its forms, including face scanning and real-time tracking, poses threats to civil liberties and individual privacy. EFF supports [bans on government use of face recognition](https://www.eff.org/document/ban-government-use-facial-recognition), [and meaningful restrictions](https://sls.eff.org/technologies/face-recognition) on use by private companies. We have [raised concerns ](https://www.eff.org/deeplinks/2025/12/ai-police-reports-year-review)about police use of generative AI technology to turn body-worn camera recordings into reports without meaningful oversight or controls. >We also oppose [government use of AI and automated tools](https://www.eff.org/press/releases/labor-unions-eff-sue-trump-administration-stop-surveillance-free-speech-online) to conduct viewpoint-based[ surveillance](https://www.eff.org/deeplinks/2026/03/government-must-not-force-companies-participate-ai-powered-surveillance) and analysis of social media because it chills free speech. EFF also investigates and [opposes](https://www.eff.org/deeplinks/2024/05/coalition-calexico-think-twice-about-reapproving-border-surveillance-tower-next) the proliferation of AI-powered technology in immigration enforcement and at the [US-Mexico border](https://www.eff.org/deeplinks/2023/03/cbp-expanding-its-surveillance-tower-program-us-mexico-border-and-were-mapping-it). Our guide [*Tackling Arbitrary Digital Surveillance in the Americas*](https://www.eff.org/wp/tackling-arbitrary-digital-surveillance-americas), compiles privacy, data protection, and access to information guarantees established within the Inter-American Human Rights System to provide concrete, actionable guidance to governments on limiting digital surveillance abuses. >Surveillance without accountability won't make us safer. The other categories include: Algorithmic Decision Making AI and Fair Use AI and NCII/Deepfakes AI and Age-Gating AI and Privacy AI and Encryption AI and Competition If you think about civil liberties, and how new technology has affected them in the past few decades, you'll see how we got to these subcategories. But are we missing any? Thanks, reddit!