Resource guide

Encode AI Is a Youth-Led Group That Pushes Laws on AI

Sneha Revanur founded Encode in 2020 to push AI laws. It publishes a funder list, no revenue figures and no date for its rename from Encode Justice.

Last updated October 22, 2026 2173-word guide By Jonathan West

Key takeaways

Sneha Revanur founded Encode in 2020 while she was a high-school student in San Jose, according to InfluenceWatch and Wikipedia.
The group now operates as Encode at encodeai.org, and its older Encode Justice site is still live, though no rename date is published.
Encode says it sponsored California SB 53, which passed, and helped pass New York's RAISE Act, which also passed.
Encode co-sponsored California SB 1047, which was vetoed, and says that bill primed the legislature for the later SB 53 law.
In April 2025, Encode joined an amicus brief in Garcia v. Character Technologies arguing the case should not be dismissed before discovery.
Encode publishes a funder list and says it takes no corporate or foreign government money, but it doesn't publish revenue figures.
Encode's tax status is unresolved, with InfluenceWatch calling it a 501(c)(3) and a separate third-party wiki calling it a 501(c)(4).

Encode is a youth-led advocacy group that pushes lawmakers to write rules for artificial intelligence (AI). Sneha Revanur founded it in 2020 while she was a high-school student in San Jose. Its homepage calls the group 'pro-human pro-AI' and says it advocates for laws.

What Encode Is and Who Started It

Encode is a youth-led advocacy group that Sneha Revanur founded in 2020. It now operates as Encode at encodeai.org, and the older Encode Justice site at encodejustice.org is still live. The group's own site publishes no date for that change, so no date is given here.

Revanur founded the group at age 15, by her own account in an interview recorded in May 2026. She says the group grew from a grassroots youth organisation into a coalition. The homepage states its mission in one line: 'We fight to make AI work for the American people.'

The homepage calls Encode a 'small advocacy team' that 'advocates for laws, builds coalitions, and influences leaders.' The site also has a profile of Revanur. For the wider field of groups working on AI policy, see our guide to who is fighting AI.

The group's first campaign opposed California Proposition 25, a 2020 ballot measure. That measure would have replaced cash bail with pretrial risk-assessment algorithms, and it failed to pass. Princeton's Prize in Race Relations says the first initiative successfully opposed Proposition 25. After that, the group broadened its focus to AI-related surveillance, disinformation and job loss.

Revanur says the group builds coalitions across unlikely allies, including conservatives, grieving parents, actors and AI safety researchers. Encode also runs AI ethics workshops. InfluenceWatch says those workshops reached over 15,000 students as of April 2023.

Who Runs Encode

The homepage lists these leaders as of October 2026:

Sneha Revanur, Founder and President
Sunny Gandhi, Co-Executive Director
Adam Billen, Co-Executive Director
Nathan Calvin, Chief Strategy Officer

The same page lists Jason Baker as Director of National Security Policy and Brian McGrail as Senior Policy Counsel. It also lists Ben Snyder and Claire Larkin as Directors of State Policy, and Seve Christian as Director of California Policy. Jesse Rothman is a Policy Advisor and Holly Branco is an Operations Associate. One press blurb on the same site calls Calvin general counsel.

A federal lobbying filing lists Encode AI Corporation at 601 13th Street NW in Washington, D.C. The filing describes it as an artificial intelligence advocacy organization registered for advocacy on 'commonsense, bipartisan artificial intelligence policy.'

How Many Members It Has

The homepage doesn't give a member or chapter count. The older figures are the group's own claims. InfluenceWatch cites, as of October 2023, 'over 600 high school and college student members in 40+ states and 30+ countries.' A January 2024 interview with the Bulletin of the Atomic Scientists reported 900 young members in 30 countries. Other third-party summaries give 1,000 or more across 40 or more states. These figures are dated differently, none is independently audited, and they don't agree with one another.

The Bills Encode Says It Shaped

The table lists the measures Encode names, including California Senate Bills (SB) 1047 and 53. Most outcomes come from the group's own account. The 'What Encode says' column carries its claims, which the sources here don't independently confirm.

Measure	Where	Outcome	What Encode says
SB 1047	California	Vetoed	Co-sponsored. Revanur says it primed the legislature and drew the governor's attention.
SB 53	California	Passed	Formally sponsored by Encode. Encode calls it the first landmark AI law, with transparency requirements and whistleblower protections.
RAISE Act	New York	Passed	Calls it the second landmark AI law, with safety transparency and 72-hour reporting of critical safety incidents. Sponsors named by Revanur: Alex Bores and Andrew Gounardes.
SB 315	Illinois	Enacted audits, per Revanur	Calls it the nation's strongest and the third landmark AI law, requiring third-party audits of frontier AI companies' safety practices.
TAKE IT DOWN Act	Federal	Law	Says it helped pass the law, which makes deepfake pornography a federal crime.
Ten-year moratorium on state AI laws	Federal	Defeated, per Encode	Credits its coalition, which it says it built and still runs. A lobbying filing records work on a letter opposing the moratorium, joined with Common Sense Media.
FY2025 National Defense Authorization Act (NDAA)	Federal	Provision claimed by Encode	Claims a provision establishing that AI must never compromise nuclear safeguards and that nuclear launch requires positive human action by the President.
FY2026 Senate defense appropriations	Federal	Report language, not enacted law	Says it secured language directing the Defense Advanced Research Projects Agency (DARPA) to prioritise AI reliability and interpretability research.

Revanur's May 2026 interview gives the most detail on SB 1047, SB 53 and the New York sponsors. She also says Illinois enacted third-party audits where California did not. Encode's own site makes the other claims in the table.

Encode AI was one of 28 signatories to a Senate letter opposing federal preemption of state AI laws. The letter was led by Americans for Responsible Innovation and the Institute for Family Studies.

In 2021, the group supported a proposed national ban on facial-recognition technology. It also ran a student week of action against facial recognition in schools with the American Civil Liberties Union (ACLU) of Massachusetts. In 2022 it lobbied the Office of Science and Technology Policy (OSTP) at the White House for an AI Bill of Rights, and it planned to back the Algorithmic Accountability Act. In May 2023 it sent an open letter to Congress and OSTP seeking more youth representation in AI policy decisions.

In January 2026, Utah State Representative Doug Fiefia introduced House Bill (H.B.) 286. The group's vice president of public policy endorsed it, saying 'we can't trust tech companies to voluntarily choose to protect our families.' The sources here don't record how the bill ended.

Its Role in Litigation

Encode has filed in two separate cases. One involves chatbot harms to a minor. The other involves OpenAI's corporate structure. The two raise different legal questions, so they are covered separately below.

Garcia v. Character Technologies

In April 2025, Encode and two other youth advocacy organisations filed an amicus brief in Garcia v. Character Technologies. The defendants argued that Character.AI's model outputs are First Amendment-protected speech, so suits based on them are barred.

Encode argued the case should not be dismissed before factual discovery, because the question is new. Its release says no United States court has held that large language model (LLM) output is protected speech. It also says First Amendment scholars treat the question as unanswered.

The judge denied the motion to dismiss on that ground. She said she was 'not prepared to hold that [LLM] output is speech,' according to FIRE's case page. Character Technologies asked for an interlocutory appeal. The sources here don't confirm how that request ended.

Adam Billen, Encode's co-executive director, argued in the release that treating the First Amendment as a complete shield against liability for harmful conduct 'could leave millions of children vulnerable for decades to come.' Readers following child-safety rules for chatbots can also read the guide to the AI Child Safety Act, the guide to the Kids Safe AI Act, and the guide to chatbot age requirements.

The OpenAI Restructuring Case

This case doesn't involve chatbot harms. In December 2024, Encode sought to file an amicus brief in Case No. 4:24-cv-04722-YGR in the Northern District of California. The brief supported the part of the plaintiffs' preliminary-injunction motion that seeks to block OpenAI's conversion into a for-profit enterprise. A LessWrong post documents the filing.

Encode's own site also says it filed an amicus brief on OpenAI's recapitalisation. It says it co-led the 'Not For Private Gain' campaign to preserve nonprofit governance. No public record shows whether the court accepted that filing.

In 2025, after that amicus brief, OpenAI subpoenaed Encode's general counsel, Nathan Calvin, at his home. A sheriff's deputy served him during dinner.

Who Funds Encode and What It Says It Turns Down

Encode publishes a funder list on its site. It doesn't publish revenue, expense or asset figures, and it doesn't publish a dollar amount for any funder. As of October 2026, the site names these funders:

Responsible Technology Youth Power Fund
Archewell Foundation
Heising-Simons Foundation
We Are Family Foundation
Future of Life Institute
Survival and Flourishing Fund
America's Promise Alliance

The list also includes individual donors and philanthropic foundations. Encode says it doesn't accept funding from corporations, foreign governments, or executives at top AI companies.

The funder list and InfluenceWatch's list don't match. InfluenceWatch says the group was initially funded primarily by the Omidyar Network. It lists the Omidyar Network, America's Promise Alliance, the We Are Family Foundation, the Princeton Prize in Race Relations and individual donations as current funders. Encode's own list doesn't name the Omidyar Network, and InfluenceWatch doesn't give dollar amounts.

A Washington Examiner investigation published on 22 April 2026 says Encode's 2023 funding came mainly from philanthropy linked to Pierre Omidyar. It also says Encode AI and the Secure AI Project rely on political donations, lobbying and public advocacy campaigns to advance their policies in Republican-led states.

InfluenceWatch reports that in January 2026 the group donated roughly $10,000 to the campaign of Utah State Representative Doug Fiefia, a Republican. Fiefia is running for Utah State Senate District 18 in November 2026. That figure comes from InfluenceWatch citing the Washington Examiner, not from a filing.

Encode AI Corporation is registered as a federal lobbying entity. Its registration appears in the Senate's lobbying disclosure database.

Its Tax Status Is Unresolved

Encode's site states no tax status, and its donation page doesn't give one. InfluenceWatch lists the group as a 501(c)(3) organisation under the U.S. tax code. It says the group is a fiscal project of Future Incubator, a joint program of the advocacy group March On and its Future Coalition.

A separate third-party wiki describes Encode as a 501(c)(4). No Form 990 turned up in the sources reviewed. Those sources also found no separate 501(c)(4) arm, such as an entity called Encode Action.

The federal lobbying filing names a distinct corporate entity, Encode AI Corporation. The sources here don't say how that entity relates to the nonprofit.

The Criticism of Encode

Encode draws criticism from a critical profile, a 2026 investigation and the founder's own admissions. Each source has its own view, so each is described below.

InfluenceWatch: A Critical Profile

InfluenceWatch is an advocacy-group database that writes from a conservative vantage point. Its Encode profile is a critical profile, and its framing is its own. The profile doesn't include Encode's rebuttals.

The profile describes Encode Justice as a youth-led activist coalition pushing for 'governance and censorship' of AI. It says the group claims 'algorithms are being used as tools of oppression.' It files the group under 'Identity Politics' and 'Tech Policy,' and it describes its funders and affiliates as left-leaning. It presents the Fiefia donation and bill endorsement as evidence of influence on state AI policy.

The Washington Examiner Investigation

A Washington Examiner investigation published on 22 April 2026 describes Encode AI as a nonprofit with 'staffing and financial ties to the Left.' It says the group helped drive the split between the Trump administration and states pursuing AI rules. It reports coordinated campaigns with the Secure AI Project in Nebraska, Tennessee and Utah.

The investigation's text, as fetched, names no critic on the record. It also shows none of the underlying records behind its claims.

What Revanur Concedes

Revanur concedes the group's early volunteer model struggled to turn energy into outcomes. That is why the group moved toward a coalition model. A skeptic's case against SB 53 says many companies already published safety plans voluntarily, so the law changed little. Revanur answers that the value lies in building state capacity.

Opposing Filings in the Garcia Case

Two groups filed positions that differ from Encode's in the same case. The Center for Democracy and Technology (CDT) and the Electronic Frontier Foundation (EFF) urged the court to weigh users' First Amendment rights carefully. NetChoice asked the court to grant Character Technologies' stay request.

Open Questions About Encode

Four basic facts about Encode remain open in the sources. Its rename date isn't published. Its tax status is contested. Its member count changes from source to source. It doesn't publish revenue, expense or asset figures.

For the group's own account of its work, start with its homepage at encodeai.org. The older domain, encodejustice.org, still carries the Encode Justice name.

Sources

Frequently asked questions

▸ Who is the founder of Encode justice?

Sneha Revanur founded the group in 2020, when she was a high-school student in San Jose. By her own account in a May 2026 interview, she was 15 when she started it. The group now operates as Encode at encodeai.org, while the older Encode Justice domain still exists.

▸ Who is Encode AI?

Encode is a youth-led advocacy group that says it advocates for laws, builds coalitions and influences leaders. Its homepage describes the group as 'pro-human pro-AI' and gives its mission as making AI work for the American people. A federal lobbying filing registers Encode AI Corporation as an artificial intelligence advocacy organization.

▸ Is Encode the same as Encode Justice?

The group now operates as Encode at encodeai.org. The older encodejustice.org site still carries the Encode Justice name. No date for the change is published.

▸ Who funds Encode?

Encode publishes a funder list that includes the Responsible Technology Youth Power Fund, the Archewell Foundation and the Future of Life Institute. It says it doesn't accept funding from corporations, foreign governments or executives at top AI companies. It doesn't publish revenue figures or dollar amounts for any funder.

▸ What bills has Encode worked on?

Encode says it sponsored California SB 53 and helped pass New York's RAISE Act, and both passed. It co-sponsored California SB 1047, which was vetoed. These outcomes come from the group's own account, and its claims about the TAKE IT DOWN Act and the moratorium fight are self-reported.

▸ How many members does Encode have?

The current homepage doesn't give member count and calls the group a 'small advocacy team.' Older figures, including over 600 members cited for October 2023, are the group's own claims. The numbers are dated differently across sources and none is independently audited.

Latest related briefings

REGULATION POLICY

Anthropic’s $1B AI Risk Bill: What It Means for People Now

A proposed law would let the government block AI models costing over $1 billion. Here’s what this means for workers, families, and students today.

Oct 21, 20265 min read

Read analysis JOBS LABOR

Anthropic AI Workforce Study Flags Shifts for Workers

Anthropic's AI workforce study signals shifting job roles and security, raising concerns for workers and families as automation accelerates in 2026.

Oct 21, 20264 min read

Read analysis REGULATION POLICY

Student Walkouts Demand AI Regulation, Safer Schools

High school students are walking out to protest gun violence and demand AI regulation, showing growing concern for safety and technology’s impact on their futures.