USPTO explores AI agents to reduce patent review backlog and examiner workload
Acting Commissioner Barry Schindler says agentic tools like Compass could cut review times from years to months by automating rote tasks and enabling human examiner interactions.
1 source · cross-referenced
- The U.S. Patent and Trademark Office is piloting AI agents to automate parts of the patent examination process, aiming to reduce a backlog of 700,000 applications.
- Acting Commissioner Barry Schindler says tools like Compass can perform formality checks, freeing examiners to focus on substantive review and human interaction.
- Schindler warns against over-reliance on AI, noting studies showing cognitive impairment from excessive agent use, and emphasizes consistency in examiner decisions.
- The agency previously deployed an enterprise AI chatbot named Scout to assist with internal research and drafting.
The U.S. Patent and Trademark Office (USPTO) is piloting agentic AI tools to automate parts of the patent examination process, including a system called Compass that performs formality checks on applications. Acting Commissioner for Patents Barry Schindler said these tools aim to reduce the current backlog of 700,000 applications and shorten review times from years to as little as one year by freeing examiners to engage in direct, human-to-human interactions with applicants.
Schindler described the patent review process as a three-step lifecycle—pre-exam, examination, and post-exam—where examiners, including PhD-level scientists, often spend time on repetitive tasks such as verifying the correctness of DNA sequences or engaging in iterative written correspondence with applicants. Compass automates these formality checks, allowing examiners to focus on substantive review and negotiation, which Schindler characterized as a human-centric process.
The agency employs around 9,000 patent examiners, many of whom are hired and placed into active roles quickly to address the backlog, leading to variability in experience levels. Schindler emphasized the need for consistency in examiner decisions, noting that agents could help newer examiners access institutional knowledge typically held by more experienced colleagues.
While advocating for the adoption of agentic tools, Schindler cautioned against over-reliance, citing research suggesting that excessive dependence on AI agents can impair cognitive skills. He stressed the importance of using agents to assist rather than replace examiner expertise, ensuring that examiners retain fundamental knowledge and judgment.
USPTO has previously deployed an enterprise AI tool named Scout, a chatbot assistant used by personnel to analyze, draft, research, and summarize content, as part of its broader AI adoption strategy.
- Nov 11, 2026 · Wired
EU proposes Digital Fairness Act to ban dark patterns and restrict addictive design features
Trust75 - Nov 11, 2026 · AI Snake Oil — Narayanan & Kapoor
New nonprofit STABL launches to study societal bottlenecks of advanced AI
Trust79 - Nov 10, 2026 · TechCrunch — AI
New York enforces AI safety law requiring registration and incident reporting for frontier labs
Trust75