Language Models & TransformersGovernance, Policy & LawSafety & Alignment

frontier model

nounAlso: frontier AI model, frontier LLM

In plain wordsOne of the most capable AI models in existence at a given time, built by a handful of leading labs at great cost, and the main target of AI safety rules.

1
General
A general-purposegeneral-purpose AI, n.: AI that can do a wide range of different tasks, like writing, coding, translating and analyzing images, rather than the one job it was built for. Today's chatbots are the main example. model at or near the leading edge of capability at a given time — typically a large language modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. or multimodal modelmultimodal large language model, n.: An AI language model that can also take in — and sometimes produce — pictures, sound or video, so you can show it a photo or speak to it, not just type. trained with the largest training computetraining compute, n.: The total number of arithmetic operations performed to train a model. It is the standard yardstick for the size of an AI training effort, and some laws use it to decide which models face extra rules. budgets by a frontier labfrontier lab, n.: One of the handful of companies building the most advanced AI models of the moment, spending billions of dollars on computing power to train them.. The boundary is relative and moves: a frontier model of one year is usually matched by mid-sized or open-weightopen-weight model, n.: An AI model whose trained internal numbers are published for anyone to download and run on their own computers, even if the data and code used to make it stay private. models within a year or two. Safety researchers and policymakers use the term for models whose new capabilities could create severe risks, which motivates dangerous-capability evaluationsdangerous capability evaluation, n.: A test of whether an AI system can do things that could cause severe harm in the wrong hands or on its own — such as helping make a bioweapon, hacking computers or copying itself., pre-deployment testing, responsible scaling policiesresponsible scaling policy, n.: A public rulebook in which an AI company promises to test each new model for dangerous abilities and to add stronger safeguards, or slow down, as its models become more capable. and incident reporting.
Every new frontier model goes through third-party evaluation before release.
2
Lawlegal
A statutory category usually defined by a compute thresholdcompute threshold, n.: A legal cutoff based on how much computation was used to train an AI model. Models trained with more than that amount face extra rules, on the theory that bigger training runs produce more capable models.. California's SB 53SB 53, prop. n.: A 2025 California law requiring developers of the most powerful AI models to publish their safety plans, report serious safety incidents to the state and protect employees who raise alarms., the Transparency in Frontier Artificial Intelligence Act (signed September 2025, in force January 2026), defines a frontier model as a foundation modelfoundation model, n.: A very large AI model, built once at great cost from vast amounts of text, images or code gathered mostly from the internet, that many products then reuse as their starting point, from chatbots to coding tools. trained using more than 10^26 integer or floating-point operations, counting compute from the original training run and from later fine-tuning, reinforcement learning or other material modifications. All frontier developers must publish transparency reports and report critical safety incidents; large frontier developers (annual revenue above $500 million) must also publish a frontier AI framework. New York's RAISE ActRAISE Act, prop. n.: A New York law requiring the largest AI developers to publish how they manage catastrophic risks and to report serious safety incidents quickly to a new state office. It takes effect in 2027., as amended in March 2026 and effective January 2027, adopts a nearly identical definition. The EU AI Act's analogous category is the general-purpose AI model with systemic riskgeneral-purpose AI model with systemic risk, n.: The EU's label for the very largest and most capable AI models, whose failure or misuse could cause harm across society, so their makers must test them for dangers and report serious incidents., presumed above 10^25 floating-point operations.

Origin

Popularized in mid-2023. Markus Anderljung and colleagues' July 2023 paper 'Frontier AI Regulation' characterized frontier models by the potential for dangerous capabilities severe enough to threaten public safety, and Anthropic, Google, Microsoft and OpenAI founded the Frontier Model ForumFrontier Model Forum, prop. n.: An industry group set up in 2023 by leading AI companies to share safety research and best practices for the most advanced AI models. the same month. The UK's November 2023 AI Safety SummitAI Safety Summit, prop. n.: A series of international meetings of governments, AI companies and researchers, begun in England in 2023, to agree on how to handle the risks and benefits of the most advanced AI. adopted 'frontier AIfrontier AI, n.: The most capable general-purpose AI systems of the moment — those at or beyond the leading edge — which governments single out for extra safety attention.', and California's SB 53 (2025) wrote 'frontier model' into law.

Usage note

Because 'frontier' is relative, fixed compute thresholds can capture models no longer at the frontier and miss efficient ones that are. Compare frontier AIfrontier AI, n.: The most capable general-purpose AI systems of the moment — those at or beyond the leading edge — which governments single out for extra safety attention., the policy umbrella term, and foundation modelfoundation model, n.: A very large AI model, built once at great cost from vast amounts of text, images or code gathered mostly from the internet, that many products then reuse as their starting point, from chatbots to coding tools., which says nothing about capability level.

Sources

Anderljung, M., Barnhart, J., Korinek, A., Leung, J., O'Keefe, C. et al. (2023). Frontier AI Regulation: Managing Emerging Risks to Public Safety. arXiv:2307.03718. arxiv.org/abs/2307.03718

Defined in law or standards

California Business and Professions Code §22757.11(i) (added by SB 53, 2025)

Mentioned in

AI governanceAI governance, n.: The rules, institutions and habits that shape how AI is built and used — from laws and treaties to technical standards and a company's own review processes., AI Safety InstituteAI Safety Institute, n.: A government agency that tests the most advanced AI models for dangerous abilities and studies how to make them safe. Britain and the United States set up the first ones in 2023., frontier AIfrontier AI, n.: The most capable general-purpose AI systems of the moment — those at or beyond the leading edge — which governments single out for extra safety attention., GPT-5GPT-5, prop. n.: OpenAI's flagship AI model from August 2025, which decides by itself whether to answer quickly or think longer. It replaced earlier models in ChatGPT and was followed by frequent upgrades., model APImodel API, n.: A web service through which programs send requests to an AI model run by a company and receive its answers, usually paying per use instead of running the model themselves., proprietary modelproprietary model, n.: An AI model kept private by the company that made it; outsiders can use it only through the company's apps or paid online services and can never download it., catastrophic riskcatastrophic risk, n.: The risk that AI could cause disaster on a massive scale — many deaths or enormous damage — through weapons, cyberattacks or systems escaping human control. Several laws now define it precisely., foundation modelfoundation model, n.: A very large AI model, built once at great cost from vast amounts of text, images or code gathered mostly from the internet, that many products then reuse as their starting point, from chatbots to coding tools., Frontier AI Safety CommitmentsFrontier AI Safety Commitments, prop. n.: Promises made in 2024 by major AI companies to publish safety plans for their most advanced models, including the danger points at which they would stop if the risks could not be controlled., frontier labfrontier lab, n.: One of the handful of companies building the most advanced AI models of the moment, spending billions of dollars on computing power to train them., GPT-4GPT-4, prop. n.: OpenAI's 2023 language model, a big step up from the one behind the first ChatGPT. It scored well on many professional and academic exams and could also read images., moatmoat, n.: A lasting business advantage that keeps competitors out, like a castle's moat; in AI, the question of whether any company can stay ahead when rivals and free models catch up quickly., parameterparameter, n.: One of the adjustable numbers inside an AI model that are tuned during training. Large language models have billions of them, and together they store what the model has learned., RAISE ActRAISE Act, prop. n.: A New York law requiring the largest AI developers to publish how they manage catastrophic risks and to report serious safety incidents quickly to a new state office. It takes effect in 2027., SB 1047SB 1047, prop. n.: A 2024 California bill that would have required makers of the biggest AI models to adopt safety plans and take care to prevent catastrophes. It passed the legislature, but the governor vetoed it., SB 53SB 53, prop. n.: A 2025 California law requiring developers of the most powerful AI models to publish their safety plans, report serious safety incidents to the state and protect employees who raise alarms., small language modelsmall language model, n.: A language model built to be compact enough to run cheaply and quickly, even on a laptop or phone, trading some knowledge and skill for speed, privacy and lower cost., state of the artstate of the art, n.: The best performance or most advanced technique currently known in a field. In AI papers, a state-of-the-art result is one that beats every earlier published result on a test., dual-use foundation modeldual-use foundation model, n.: A US legal term for a general-purpose AI model that serves everyday uses but is capable enough that it could help someone build weapons, mount cyberattacks or evade human control., GPU poorGPU poor, adj.: Slang for researchers, startups or countries without access to the huge numbers of expensive chips needed to build the most advanced AI, as opposed to the GPU-rich big labs., small AIsmall AI, n.: Cheap, compact AI tools built for one practical job — like advising farmers or screening patients — that poorer countries can afford to run, instead of giant all-purpose models.