Meredith Whittaker

Meredith Whittaker is the President of the Signal Foundation (September 2022–; board member since 2020) and co-founder and chief advisor of the AI Now Institute — the New York City research-policy institute she built with Kate Crawford in November 2017 as the first university institute dedicated to the social implications of AI (see Person entry). She is tracked here as a Voice because her sustained named public output — the Signal presidency as an argument-in-action about privacy infrastructure at scale; the AI Now co-founding and the labor-and-power-concentration research agenda it carried; the two congressional testimonies on AI (2019) and facial recognition (2020); the role as senior advisor on AI to FTC Chair Lina Khan (November 2021–September 2022); the core organizing role in the November 2018 Google Walkout for Real Change; and the sustained public-voice record articulating that AI is inseparable from surveillance capitalism — collectively carries the working frame that AI is a technology of concentrated corporate power built on surveillance infrastructure, whose harms fall now on marginalized communities, and whose remediation requires structural political action (privacy legislation, workplace organizing, antitrust) rather than technical ethics frameworks.

Voice anchors

The Voice anchors four registers the corpus's voices slice had previously left underweight.

• The privacy-infrastructure-at-scale voice anchor entirely. No other Voice in the corpus operates global communications infrastructure. Whittaker's Signal presidency is not primarily an advocacy position — it is an ongoing structural argument, expressed operationally: the Signal Foundation runs the world's most widely deployed privacy-focused end-to-end encrypted messaging platform as a nonprofit with no advertising, no data monetization, and no compromise on encryption, used by journalists, human rights defenders, political organizers, and tens of millions of users globally. The decisions Signal makes about feature design, business model, and encryption policy are themselves a form of public voice — and Whittaker has articulated them explicitly, arguing that private, secure communication is not a consumer preference but foundational infrastructure for political organizing and democratic life. No other corpus voice closes this anchor.

• The professional-tech-worker-collective-action voice anchor. The corpus's labor-organizing voices are concentrated in the gig-worker and content-moderation registers: Daniel Motaung (Sama / Facebook content-moderation whistleblowing), James Farrar (Uber gig-worker strategic litigation). Whittaker's core-organizer role in the November 2018 Google Walkout for Real Change — in which approximately 20,000 Google employees and contractors globally walked off the job, the largest coordinated tech-worker collective action in US history to that point — anchors a structurally distinct register: professional software engineers and researchers at one of the world's largest AI companies exercising collective voice on AI procurement and corporate governance. Her subsequent retaliation by Google — reportedly told to abandon her AI ethics work, departing in July 2019 — and her public departure commitment to AI ethics work and organizing give the walkout narrative a clear continuity arc through to AI Now and Signal. The Voice closes the professional-tech-worker-organizing voice sub-type entirely.

• The AI-as-surveillance-capitalism voice anchor. Whittaker's signature public frame — "AI is fundamentally a surveillance technology" — occupies a distinct slot from the corpus's other accountability voices. Kate Crawford anchors the extractive/material-infrastructure register (minerals, labour, energy, data costs). Timnit Gebru anchors the LLM-critique and algorithmic-accountability research register. Joy Buolamwini anchors the facial-recognition-bias and participatory-audit register. Whittaker's specific contribution is the structural business-model argument: AI is inseparable from the surveillance-advertising industry that funded the data infrastructure on which it runs — "the Venn diagram is a circle" — and the only meaningful accountability path runs through privacy legislation, antitrust to break the market concentration that makes AI buildable only by a handful of corporations, and workplace organizing to give workers inside AI companies collective voice. Without this Voice entry, the corpus's clearest structural framing of AI as surveillance-capitalism infrastructure had no corresponding voice anchor.

• The US regulatory-channel voice anchor. Whittaker's two congressional testimonies — on AI's societal implications (2019) and on facial recognition accuracy and transparency (2020) — and her FTC senior-advisor-on-AI role under Chair Lina Khan (November 2021–September 2022) make her the corpus's most direct voice-to-US-regulator channel. Other corpus voices reach US regulatory audiences through research reports and policy submissions; Whittaker's channel was direct advisory and testimony, giving her voice a throughput into active enforcement decisions that most of the corpus's activist-side voices reach only indirectly. TIME named her among the 100 most influential people in AI in 2023, anchoring the international-influence register through which her voice carries into mainstream policy and business audiences beyond the core AI-research community.

Signature framings

Four framings recur across Whittaker's public output and have done the most to install her register into the make-AI-good movement.

• "AI is fundamentally a surveillance technology." Whittaker's central public argument — stated most directly in her September 2023 TechCrunch interview — holds that AI and the surveillance-advertising business model built by Google and Meta since the late 1990s are structurally inseparable: the mass data collection that surveillance advertising required is the training data on which modern AI systems run, and "the Venn diagram is a circle" when comparing AI development to the surveillance-advertising business model. Where Crawford's extractive-infrastructure critique centres AI's material costs (minerals, labour, energy), Whittaker's business-model critique centres AI's informational costs — what surveillance capitalism required to build the data infrastructure, and whose data and privacy it cost. The argument is politically consequential: it reframes privacy legislation — long positioned as a consumer-protection concern — as the most direct structural lever available to interrupt AI's data-collection dependencies.

• Present harms over hypothetical AI futures; "tech-washing precarious human labor." Whittaker has consistently argued that AGI-risk and "AI doomerism" narratives function as a corporate distraction mechanism, shifting regulatory and public attention from the present harms AI systems inflict on "low-wage workers, people who are historically marginalized, Black people, women, disabled people" — deployed with very little accountability, almost no testing, and backed by extremely exaggerated claims that function as corporate marketing. She extends this to AI annotation labor: thousands of workers annotating training data and providing reinforcement learning with human feedback for minimal pay are being "tech-washed" — their precarious human labor rendered invisible beneath the automated appearance of the AI systems they build. This framing links her Signal-side privacy argument to the AI Now–side labor-and-power-concentration argument into a single structural critique.

• Structural accountability over technical ethics. Whittaker's Slate 2023 argument that accountability requires workplace organizing, structural corporate separation, and community-level resistance rather than technical ethics audits or voluntary industry self-regulation is the corpus's clearest articulation of the structural-vs-technical accountability split. Where industry-side AI ethics frameworks treat accountability as a technical problem (better benchmarks, bias audits, explainability requirements), Whittaker's frame treats it as a political-economy problem requiring structural power shifts — the same argument the AI Now Institute's 2025 Artificial Power landscape report carries forward institutionally. The two threads — the research-policy register (AI Now) and the privacy-infrastructure register (Signal) — converge here: Signal's nonprofit, no-monetization, strong-encryption model is not merely a product decision but a structural argument about what private communications infrastructure must look like to serve political organizing rather than surveillance.

• Power concentration as the core AI problem. Only a handful of corporations have the computational resources, training data, and capital required to build frontier AI systems — a concentration that makes AI governance structurally inseparable from antitrust. Whittaker has carried this framing consistently from the AI Now period through her FTC advisory role (during Lina Khan's renewed antitrust enforcement interest in tech markets) and into the Signal presidency. The November 2023 Harvard Business Review podcast illustrates the cross-audience reach of this argument — carrying the AI-as-power-concentration frame into business-leadership audiences that the movement's activist-side voices rarely reach directly.

Public output and venues

Whittaker's public-facing work runs through four overlapping channels.

• Signal Foundation presidency. Her primary current platform is Signal itself — both the platform's operational governance (feature decisions, encryption commitments, business-model choices) and the public-voice channel it provides. The September 2022 founding announcement framed her presidency as combining Signal's privacy infrastructure with her AI accountability research background. Subsequent Signal communications — on feature decisions, on law-enforcement access requests, on encryption legislation — carry her voice at a scale no research publication can match.

• Congressional testimony and FTC advisory work. Whittaker has appeared before Congress twice — on AI's societal implications (2019) and on facial recognition accuracy and transparency (2020) — and served as senior advisor on AI to FTC Chair Lina Khan from November 2021 to September 2022. This direct advisory and testimony record is the corpus's most direct voice-to-US-regulator channel, giving her work a throughput into active enforcement agendas not available to research-only voices.

• AI Now Institute research and policy output. As co-founder and faculty director of the AI Now Institute from 2017 to 2022, Whittaker shaped the research agenda that produced the Institute's annual landscape report series and the April 2019 Discriminating Systems: Gender, Race, and Power in AI report — co-authored with Sarah Myers West and Kate Crawford. The Discriminating Systems report argued that the AI workforce diversity crisis and AI system bias are the same structural problem: the pipeline framing (fix diversity by adding underrepresented groups to existing AI pipelines) fails because the pipeline reproduces the power structures that produce biased systems. As current chief advisor, her influence on the Institute's research direction continues into the post-2022 phase, including the 2025 Artificial Power landscape report's power-concentration framing.

• Public interview and keynote circuit. Whittaker carries a sustained named-interview and keynote register across academic, policy, and mainstream-media venues. The May 2023 Slate interview, the September 2023 TechCrunch interview, and the November 2023 Harvard Business Review podcast illustrate her cross-audience reach: from tech-specialist to policy-focused to business-leadership audiences. The TIME 100 Most Influential People in AI 2023 recognition anchors the international-public-influence register.

Organisational vehicles

Whittaker's public output runs through two principal organisational vehicles. The Signal Foundation — the San Francisco-based nonprofit she has led as president since September 2022 — is her primary current platform, combining operational governance of global privacy infrastructure with a sustained public-voice register on encryption, surveillance, and the political economy of private communications. The AI Now Institute — the New York City research-policy nonprofit she co-founded with Kate Crawford in November 2017 — remains her research-and-policy vehicle through her continuing role as chief advisor; the Institute's 2025 Artificial Power framing carries forward the power-concentration argument she built into its founding research agenda. A third earlier vehicle — her thirteen years at Google (2006–July 2019), through which she founded Google's Open Research group, co-organized the Project Maven petition, and co-organized the Google Walkout for Real Change — is the origin context for her departure and the organizing commitment she carried into AI Now's accelerated independence phase.

Why this is a Voice entry

A Voice entry is created here, rather than additional structure on the Person entry, because Whittaker's public-facing output is itself the load-bearing object the corpus needs to track: the Signal presidency as an argument-in-action about privacy infrastructure at scale; the core-organizing role in the November 2018 Google Walkout for Real Change (approximately 20,000 participants globally); the Discriminating Systems co-authorship linking AI workforce diversity deficits to AI system bias; the two congressional testimonies and the FTC senior-advisory role giving her voice a direct throughput into US regulatory processes; the sustained public-interview and keynote record articulating that AI is inseparable from the surveillance-capitalism business model; and the TIME 100 Most Influential in AI 2023 recognition anchoring her international-public-influence register. The corpus's accountability voices had strong coverage on the algorithmic-bias-and-harm (Gebru, Buolamwini, Noble, Benjamin) and extractive-material-infrastructure (Crawford) registers but no voice anchoring the AI-as-surveillance-capitalism structural register, the professional-tech-worker-organizing register, or the global-privacy-infrastructure-at-scale register — this entry closes all three. Affiliation, training, and biographical detail are recorded on the linked Person entry per the corpus's Person/Voice split.