社会ガイド
USPTO Guidance on AI-Assisted Inventions
The USPTO’s November 2025 guidance says the same inventorship standard applies whether or not AI was used in the inventive process.
このページでは3 分で読めます
概要
A person may use AI as a tool, but only natural persons can be named as inventors under the guidance; patentability still depends on the ordinary requirements.
ディープダイブ
The United States Patent and Trademark Office issued revised inventorship guidance in November 2025. It clarifies that the same legal standard for determining inventorship applies to all inventions, whether or not an AI system was involved. The earlier February 2024 guidance is archived by the USPTO and should not be treated as the current policy. The updated guidance centers the analysis on human inventorship under existing law. An AI system is not a natural person and cannot be named as an inventor. That does not mean that an invention created with AI assistance is automatically unpatentable. A person who uses AI may qualify as an inventor if that person made the contribution required under the ordinary inventorship standard. The legal inquiry concerns the human contribution to conception, not how impressive or autonomous the software appears. Inventorship is separate from other patent questions. A filing still must satisfy applicable requirements such as eligible subject matter, novelty, nonobviousness, and adequate disclosure. The 2025 guidance addresses inventorship; it does not guarantee a patent, resolve ownership of AI outputs, decide copyright questions, or determine whether a particular AI-generated result is itself patentable. Those issues require separate analysis. The USPTO FAQ states that the guidance does not create a heightened inventorship standard for AI-assisted inventions and does not impose an additional duty to disclose the extent of AI use beyond existing requirements. Applicants and practitioners remain responsible for existing duties and for the accuracy of filings. In practice, teams should preserve evidence of how a concept developed and identify the people whose contributions meet the applicable standard. This overview is educational, not legal advice about a particular application.
戦略的影響
リスクと安全性
AI による壊滅的な被害も日常的な被害も、誰がリスクを理解し、誰が行動できるかにかかっています。
より明確な判決
国民と専門家のリテラシーは、強力な安全政策が政治的に可能かどうかを左右します。
誇大広告を打ち破る
明確な説明は、誇大広告、研究室の PR、曖昧な倫理劇場に囚われることを減らします。
The Future of USPTO Guidance on AI-Assisted Inventions
AI tools will continue to contribute to research and design, while inventorship remains a human legal determination under the current USPTO guidance. The Office may update examples or FAQs as practice develops. Teams should date their internal guidance, retain development records, and check current USPTO materials before filing. Separate inventorship from unresolved questions about ownership, patent eligibility, and rights in AI outputs. Revisit role assignments when an invention team or tool changes. Assign an owner for the inventorship review and keep the supporting reasoning with the application file.
現実世界の実装
An engineer documents the human conception and development work on an invention while using a model to explore possible structures.
A patent team reviews who contributed to conception rather than listing a generative model as a joint inventor.
An applicant evaluates novelty, nonobviousness, and disclosure separately from the question of who qualifies as an inventor.
A company updates its invention disclosure form to capture human contributions without imposing an AI-use disclosure duty that the USPTO guidance does not create.
リスクとガードレール
能力が複雑になる一方で、実存的なリスクを SF として扱います。
高度な自律性の下での調整による表面製品の安全性を混乱させる。
英語以外や専門家ではない聴衆には、低品質の情報源しか提供されません。
実装ロードマップ
製品の危害、誤使用、制御不能/調整不良のリスクを分離します。
どのような証拠がタイムラインと重大度についてのあなたの見方を変えるかを尋ねてください。
マーケティング上の主張よりも、一次情報源と具体的な評価を優先します。
意識だけでなく、キャリア、政策、資金、スキルなど、行動経路を 1 つ特定します。
探検を続けましょう
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the USPTO Guidance on AI-Assisted Inventions quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
よくある質問
What is USPTO Guidance on AI-Assisted Inventions?
The USPTO’s November 2025 guidance says the same inventorship standard applies whether or not AI was used in the inventive process. A person may use AI as a tool, but only natural persons can be named as inventors under the guidance; patentability still depends on the ordinary requirements.
What did the USPTO clarify in its November 2025 AI inventorship guidance?
The current USPTO guidance says the same standard applies regardless of whether AI was used.
Who may be named as an inventor under the USPTO guidance?
The USPTO treats inventorship as a human legal determination; an AI system is not a natural person.
Does using AI automatically make an invention unpatentable?
The USPTO states AI-assisted inventions are not categorically unpatentable; ordinary patent requirements remain.
Which contribution is central to the current inventorship analysis?
The guidance focuses on human contribution under the same inventorship standard.
How should a team treat the February 2024 USPTO guidance after the 2025 update?
The USPTO identifies the November 2025 guidance as current and archives the earlier version.
学び続ける
関連ガイド
このトピックのために選ばれたその他のガイド