
Executive Director of Partnership for Assurance, Credibility, and Trust on AI @pactai_org Working on the new economy of trustworthy AI
Bay Area
Joined March 2023
- Tweets78
- Following189
- Followers95
- Likes134
The report is out: California just released recommendations on accelerating the implementation of the first-ever independent AI oversight framework passed into law.
It includes so much of what PACT AI members recommended to the state, as the organizations doing the independent AI evaluation work these laws will impact. Huge thanks to all our members at @pactai_org whose insights and expertise have impacted a first-in-nation AI oversight framework. Some key points from PACT in the final report include:
✅ Moving toward mandates to establish a robust IVO ecosystem in a way that balances the need for action now with the practical realities of an evolving technology and evaluation science.
✅ Defining minimum conditions for credible, independent AI evaluations and audits.
✅ Growing a competitive marketplace for innovation.
The implications of this report extend well beyond California. It will no doubt be a blueprint for how other states and policymakers can jumpstart the AI assurance ecosystem that we need to meet the demand in the economy.
Read California’s full report here: gov.ca.gov/2026/11/18/govern…
1
18
Exciting news via @POLITICOPro: just a couple months after @pactai_org launched, we’re already leveraging the deep AI assurance expertise of our members to advise California’s implementation of its independent verification law, and jumpstart a strong and effective IVO ecosystem.
This is exactly what PACT AI was designed to do. Consumers and companies can’t afford to simply take it on faith that the AI models and systems underpinning the economy are working as designed. We represent the independent verification organizations (IVOs) doing the technical work to verify that AI performs safely, securably and reliably, as well as the businesses that need that assurance in order to adopt this technology.
We’ve seen this pattern in other high-stakes industries: financial markets run on independent audit; autos, airplanes, electrical products run on independent safety testing. We need a professional assurance sector for AI, too. One that gives buyers, insurers and the public a basis for trust that does not depend on the seller's word. That trust is what allowed those industries to grow and allowed our society to reap the benefits of the technology.
Here are the technical recommendations PACT members provided to California as they lay the groundwork for this ecosystem:
(1) Defining minimum conditions for credible, independent AI evaluations and audits. PACT recommends four conditions for trustworthy evaluations: independence, expertise in the specific risks being tested, access to the systems and evidence needed, and thorough testing methods that keep up with the technology.
(2) Moving toward mandates to establish a robust IVO ecosystem. Waiting until the frontier AI IVO ecosystem is fully "ready" is a chicken-and-egg problem: the market won't grow as quickly as it can without clearer demand. PACT AI outlines several ways the state can shift from voluntary compliance toward mandatory requirements in a way that balances the need for action now with the practical realities of an evolving technology and evaluation science.
(3) Growing a competitive marketplace for innovation. The current risk to frontier AI IVOs’ independence is structural. Since there are currently only four to six frontier developers, an IVO focused primarily on working with frontier developers will likely get more than 20% of its revenue from a single developer. To address this and other structural risks around revenue concentration, the threat of retaliation, and protecting evaluator independence, PACT AI identifies practical steps the state can take to nurture a quality, competitive marketplace of evaluators.
PACT AI members are already doing the hard work of verifying, testing, and auditing AI systems, and the coalition’s recommendations reflect their deep expertise and shared goal of ensuring trust in AI scales alongside innovation.
You can read our letter here: pactai.org/resources-news/pa…
3
29
Bri Treece retweeted
“A big-name push for AI audits”: @BrendanBordelon has the first look at our new Executive Council in @politico Decoded this morning.
“Fathom was the driving force behind California’s passage this year of SB 813, which requires the state to set standards for outside groups that evaluate AI safety risks. It now wants to spread that to other states, countries and Washington DC.”
Trust in AI gets earned in the rooms where decisions are made, not stamped on at the end. PACT AI’s VP of AI Assurance Provider Services, Graeme Finley spent twenty years in those rooms, and his read on what keeps evaluators independent once they're inside is worth the time for anyone building or deploying these systems.
"Embedded evaluators" is a new idea in AI governance, but embedded oversight has been around for decades. Graeme Finley, our VP of AI Assurance Provider Services, spent 20+ years with a badge and a desk inside large state government IT projects. Here's what that work teaches.
1
40
I talked to @POLITICOPro about @pactai_org and the nascent AI assurance ecosystem. Right now, only a handful of organizations are doing independent evaluations of AI systems, and we’re going to need a lot more working on the frontier and the novel risks that are arising.
That’s why we’re building the trust infrastructure this field needs to scale, with assurance providers, deployers, and insurers at the same table.
Bri Treece retweeted
On the floor at #IAPPpsr26 in Seattle, where AI assurance is a big part of the conversation this week.
At 11:15 AM PT today, @pactai_org's Graeme Finley joins Ashley Casovan, Miranda Bogen and Craig Shank in Room 4C-2 for: "What is AI Assurance? Navigating a Rapidly Evolving Ecosystem."
If you’re here, come see Graeme speak or connect with him after the session.
What a night! Thank you to everyone who joined us in DC on the Wild Days rooftop to celebrate the launch of @pactai_org, and a huge thank you to the PACT AI team who made it happen.
The conversation around AI assurance has moved fast. We're no longer asking whether we need it, but how we do it well. That's why PACT AI exists: to bring together the deployers, assurance providers, insurers, and researchers building the infrastructure for a thriving AI assurance ecosystem.
It was great to see so many of you in one room.
If you want to learn more or connect with our team, check out pactai.org.
Exciting news for @pactai_org today: Rob Reich has helped inform how PACT AI thinks about the assurance ecosystem as a member of our Advisory Council. Now, he and Dr. Gillian Hadfield, an architect of the IVO framework, are helping bring it to life in California.
Congratulations to Rob Reich, @PACTAI_org Advisory Council member, who was named to Governor Newsom’s panel of AI experts that will help bring the AI assurance ecosystem to life in California.
This builds on news from last week that California is speeding up implementation of its new laws to build out the AI assurance ecosystem –– establishing a framework for independent verification organizations (IVOs) that assess AI models and systems for safety and risk, and create a state registry for AI auditors with standards for their independence, transparency, and integrity.
Thank you to the four experts named to this group: Rob Reich, Gillian Hadfield, Alondra Nelson, and Jason Goldman for this important work advising the establishment of California’s AI assurance ecosystem.
3
211
Big news for PACT AI out of California: Governor Newsom just signed an executive order to accelerate the implementation of SB 813: the creation of an independent AI verification ecosystem.
BREAKING: @CAGovernor Gavin Newsom just issued an EXECUTIVE ORDER directing actions to address the dangers of recent AI incidents.
The order accelerates California’s new law establishing first-in-the-nation independent oversight of AI companies and safety checks and advances the creation of an “AI kill switch.”
More to come — watch this space!
It also speeds up the clock: the criteria for what organizations could be IVOs are now due May 1, 2027, instead of January 2028.
1
1
31
Bri Treece retweeted
Delegating work to AI does not remove the responsibility to govern it.
If an AI agent can move money, access sensitive information, or make a commitment on your company’s behalf, someone needs to answer three questions:
✅ Who gave it that authority?
✅ What limits apply?
✅ When must a person intervene?
As a founder and board director, I want those answers supported by evidence that leaders can examine and act on.
That is why I’m pleased to share that Alpha has joined PACT AI @pactai_org, the Partnership for Assurance, Credibility, and Trust on AI.
Technical evaluations help us understand how a system behaves. Boards and executives also need to understand whether it is appropriate for their organization to deploy that system, for a particular purpose, with particular permissions and oversight.
Connecting those questions is central to what we are building at Alpha.
Thank you to @BriTreece_ and the PACT AI coalition for bringing enterprises, independent experts, insurers, and civil society together around credible AI assurance. We look forward to contributing our infrastructure work, research, and boardroom perspective.
The more responsibility we delegate to AI, the more clearly we need to establish human accountability.
#AIGovernance
#PactAI
#AgenticEnterprise
Read the press release here: lnkd.in/gHmzp9tH
As AI agents gain authority to act, how do organizations prove they are governing them responsibly?
Alpha has joined the Partnership for Assurance, Credibility, and Trust on AI (PACT AI) @pactai_org to help advance verifiable governance for the agentic enterprise.
The questions are practical:
✅ Who authorized an agent to act?
✅ What limits apply?
✅ Who is accountable?
✅ What evidence shows that its controls are working?
Answering them requires technical assurance and enterprise governance to work together. Alpha brings governance infrastructure, research from the Alpha Institute for AI Governance, and the boardroom experience of the Alpha Council to this effort.
We look forward to working with @BriTreece_ and the PACT AI coalition to connect assurance findings with the decisions boards and executives make about deploying AI, expanding its authority, and intervening when necessary.
Governance is Alpha.
#PactAI
#AIGovernance
Read the announcement:
lnkd.in/gNkhUC7J
Bri Treece retweeted
As AI agents gain authority to act, how do organizations prove they are governing them responsibly?
Alpha has joined the Partnership for Assurance, Credibility, and Trust on AI (PACT AI) @pactai_org to help advance verifiable governance for the agentic enterprise.
The questions are practical:
✅ Who authorized an agent to act?
✅ What limits apply?
✅ Who is accountable?
✅ What evidence shows that its controls are working?
Answering them requires technical assurance and enterprise governance to work together. Alpha brings governance infrastructure, research from the Alpha Institute for AI Governance, and the boardroom experience of the Alpha Council to this effort.
We look forward to working with @BriTreece_ and the PACT AI coalition to connect assurance findings with the decisions boards and executives make about deploying AI, expanding its authority, and intervening when necessary.
Governance is Alpha.
#PactAI
#AIGovernance
Read the announcement:
lnkd.in/gNkhUC7J
The AI innovation frontier is no longer just about advancing model capability. It’s about advancing AI assurance.
There's a lot to unpack from this weekend: Anthropic CEO Dario Amodei called for the industry to pace frontier development, specifically calling for embedding independent third-party evaluators within the labs themselves. Sam Altman and Elon Musk quickly agreed. If nothing else, these developments make it clear that being able to verify AI models and systems are working as intended is the most essential work in AI right now.
So, how do we execute third-party assurance well? How can we trust AND verify that the models powering our economy are safe, secure, and working as intended? Just ask our @pactai_org members. They know that a slowdown alone doesn't build trust.
PACT AI's assurance providers, enterprise deployers, technical experts, and civil society leaders are already paving the way, developing the latest science, methods, and tools to drive clarity and accountability. Together, we're building a robust AI assurance ecosystem so that trust in the AI economy can scale as fast as innovation, whatever pace the frontier sets.
This work is more important than ever, and we're just getting started.
📣 Members: Join us for a briefing on this today at 4:30PM ET (info in your email) 📣
Interested in learning more ⬇️
Visit our website pactai.org to learn more and to connect.
1
30
Bri Treece retweeted
Independent evaluation of AI may be new to governments' policy agendas, but the model behind it is not.
Governments have delegated technical validation for decades: standards bodies, conformity assessment schemes, financial auditors. 🧵
Bri Treece retweeted
The lesson of the Hugging Face incident is not just that 700 AI agents went rogue. It's that the people who investigated afterward were there by invitation, on the company's terms, with no legal category that says who is qualified to do that work or what access they get. Yesterday California took the biggest step yet toward changing that and building the independent verification sector we need, and I'm excited to see it signed into law. Governor Newsom signed SB 813, which directs the state to designate independent verification organizations (IVOs), outside experts qualified to assess the risks of AI models. He also signed AB 1405, Assemblymember Rebecca Bauer-Kahan's bill creating the state registry AI auditors (including IVOs) have to join before they can conduct an audit.
OpenAI endorsed both bills. It would prefer independent technical assessments required at the federal level, where standards for who is qualified and what access they get could be set consistently. But absent federal action, it said, California can help set the rules of the road for a national system.
A state can only do so much of that on its own. The FRONTIER Act, the bipartisan bill from @JayObernolte and @RepLoriTrahan, would do it nationally, licensing independent verifiers and requiring the largest developers to retain one with the access to do the work.
Thank you to @SenMcNerney and @CAgovernor, and to @AndrewFATHOM, @deanwball and everyone at @Fathom_org, which sponsored the bill, for getting this over the line. There's a lot more to do, in other states and in Congress, and we'll keep at it. It’s critical that we start building an independent oversight ecosystem.
buff.ly/V5hzUAo























