Post-webinar Q&A: Smart filters in patent review

Thank you to everyone who attended our recent webinar on Smart filters and a special thank you to Hannu Huotari, Director, Strategic IP at ASM, who joined us as a guest speaker and presented a real-life use case. We had some great discussions, and we appreciated all the insightful questions from the audience. Below is a summary of the Q&A session, now enhanced with additional details about Smart filters' functionality, reliability, and best-use scenarios.

Q: How long documents can Ask AI and Smart filters handle?

A: Ask AI can handle input documents up to 200,000 tokens, which corresponds to about 200–300 pages. Since the system analyzes one document at a time, almost all answers are based on the entire document – except in the rare cases where the document exceeds this limit.

Q: How can I view a long answer?

A: There are a few ways to see the entire answer:

  • Hover over the answer to see a popup with the full response.
  • Click on the answer to open the full conversation in the Ask AI drawer.
  • Export to Excel, where all questions and answers for each document will be visible.
  • Zoom out to make text smaller

Q: How long are filters kept active? Are they tied to specific queries?

A: A Smart filter remains available if at least one document on the current page has an answer to that question. Filters stay active within the same search case or collection but do not carry over to different cases or collections.

Q: Would reversing the order of two related questions produce the same shortlist of results?

A: In principle, yes, but in practice, there may be slight differences.

Q: IPRally used to show the number of available credits when making Ask AI questions. What happened to that?

A: IPRally no longer uses the credits system.

If you don’t see the amount of credits anymore, it means your company already has Unlimited Ask AI enabled, and you can ask any number of Ask AI questions.

If your usage under the new system becomes excessive to the point of impacting system performance, we may introduce soft limits. If that happens, we will notify you well in advance.

Q: Do Smart filters miss relevant results that Boolean searches can find?

A: Boolean searches match exact words or word fragments in documents. Smart filters, on the other hand, analyze the actual meaning of the document. This means:

  • Boolean searches may find documents with specific words, even if they are not relevant.
  • Smart filters find documents that conceptually match your question, even if they use different wording. They can also occasionally miss details or misinterpret documents.

The methods are hard to compare directly, because it depends on the exact queries and prompts. 

Best practice: Use Smart filters for feature-level analysis in novelty, invalidity, and state-of-the-art searches, while using Boolean searches for keyword-based filtering.

Q: How do Smart filters handle non-English patent documents?

A: Ask AI and Smart filters only considers the English translation of documents. Similarly, IPRally’s search features are also based on English translations.

Q: Is there an advantage to using Smart filters in addition to free-text search?

A: Yes. IPRally’s AI free-text search helps narrow down millions of patents to a few hundred relevant ones. However, it is not always precise in matching exact technical details.

Smart filters go deeper, analyzing the content of those top results to refine them further – so you get e.g. from 500 documents to the top 20 most relevant ones.

Think of it this way:

  • IPRally's AI search helps you find broadly relevant patents.
  • Smart filters help you review them in detail by filtering based on specific technical features.

Q: Can Smart filters recognize and filter based on term definitions in disclosures?

A: Yes. Since Ask AI sees the entire patent document, including term definitions, it can filter results based on them.

Q: How long does it take to apply Smart filters to 1,000 patents?

A: Typically a few minutes, but processing time depends on how many users are running Ask AI at the same time. If you experience long wait times, let us know.

Q: How does Smart filters handle unclear questions?

A: The AI makes its best guess based on the question’s wording. You’ll get better results by using clear, well-defined Yes/No questions.

Example of a good Smart filter question:
"Does the document disclose a piezoelectric transducer suitable for pressure sensing?"

Avoid vague questions like:
"Is this patent relevant to my project?"

Q: What are the AI and Boolean scores that can be exported?

A: These are numerical scores (0–100) that indicate how closely a document matches the search query:

  • AI Score for AI-based searches.
  • Boolean Score for Boolean-based searches.

Q: Can multi-patent Ask AI produce a less detailed answer than single-patent Ask AI?

A: If you ask the same questions in the same order, the answers should be identical – regardless of whether you ask them one document at a time or in bulk.

Q: What does “Maybe” mean in Smart filters?

A: The Maybe option has been added to give the AI the ability to express uncertainty. It means that the AI couldn’t make a definitive Yes or No decision. 

Why?

  • The question might be too complex or vague.
  • The patent might lack clear information.

Solution: Try asking a more specific Yes/No question to get a clearer answer.

Q: How does Smart filters handle patent family members?

A: Search results in IPRally are grouped based on basic family logic, prioritizing documents in the following order:

  1. US 
  2. EPO
  3. PCT
  4. Other native English-language documents (GB, AU)
  5. Remaining global documents

Smart filters always analyze and provide answers based on the primary family member shown in the results list. If you want to ask questions about a different family member, you can open the family list and start a conversation about that specific document. However, this will not affect the Smart filter applied to the original search results.

Q: How reliable are Smart filter results?

A: Smart filters rely on LLMs, which are not perfect. While they provide valuable insights, always double-check the results, as:

  • AI may misinterpret details or misunderstand a question.
  • Shorter patents tend to have more reliable answers than longer ones (e.g., pharmaceutical patents).

The reliability goes up as new more powerful models are taken into use.

Q: Can Smart filters analyze image-based data?

A: No. Smart filters rely only on text and do not analyze images. However, they can process tables and chemical formulas if they are presented in text format rather than images.

Q:  Where can I find the webinar recording? ‍

A: You can find the on-demand webinar here.

Can I try IPRally?

Yes, we’re happy to offer you 3 days of free access to IPRally, including Smart filters. No costs, no commitments – just the opportunity to see how it can change the way you search and review.

‍

Read more

Thank you to everyone who joined No more drafting in the dark: prior-art-aware patent drafting with IPRally Protection. Across our EU and US sessions, you sent us more than 80 questions, from file formats and drawings to claim clarity and US disclosure duties. We have grouped the answers by theme and merged similar questions. Some were answered live or in the chat during the sessions, and we include those here too.

Webinar Q&A follow-up: your questions about IPRally Intelligence answered

Your questions on IPRally Intelligence, answered: from working with larger datasets to verifying results at paragraph level.

IPRally is SOC 2 Type II compliant

IPRally has completed its SOC 2 Type II examination. Advantage Partners, our independent auditor, tested our security controls in operation from March 4 to June 4, 2026, and issued an unqualified opinion: no exceptions noted on any control tested.

When systems start talking: IPRally's journey toward a virtual IP department

Something new is happening: people talk to systems in plain language, systems talk back, and systems have started talking to each other the same way. As AI climbs from the small decisions to the large ones, the question for IP work is no longer whether it gets automated, but how to stay in control once it is. Co-founder and CPO, Sakari Arvela, lays out where this leads – through IPRally's new API and upcoming MCP support toward a virtual IP department that's always on – and why humans end up more important in that picture, not less.

Search, Intelligence, Protection: the three pillars of IPRally’s next chapter

For nearly a decade IPRally has been obsessed with getting one task right: patent search. Having built the best patent search in the world, we’re now reorganizing around three pillars – Search, Intelligence, and Protection – the newest of which makes drafting prior-art-aware from the first sentence. Co-founder and CPO Sakari Arvela maps out the next chapter, and why Protection is the shift he's most personally invested in.

NGen selects IPRally to bring AI-native patent intelligence to Canada's advanced manufacturers

After evaluating a range of tools on functionality, usability, and cost, Canada's Global Innovation Cluster for Advanced Manufacturing chose IPRally to power its Patent Search Adoption Tool Pilot Program.

IPRally and RPX Corporation announce strategic collaboration, cross-linking AI-powered patent search with RPX Empower

IPRally today announces a strategic collaboration with RPX Corporation, the global patent risk management service provider, that further expands the capabilities of the RPX Empower platform and provides seamless access to business critical complementary information for IPRally users.

Macro meets micro: IPRally's next chapter builds intelligence on search

Excellence in patent search was always the foundation. But foundations are meant to be built on. After years of staying focused – and saying no to a lot – IPRally is now ready to bring the same rigor and precision that redefined search to the bigger, strategic questions that businesses make decisions on. CCO Andreas Cehlinder explains why now, and what comes next.

What does quality actually mean for automated patent search?

When agentic AI tools entered the patent search market, quality became the topic everyone wanted to talk about, and one of the hardest to discuss honestly. Impressive-looking outputs are easy to generate, but outputs you can actually trust take a different kind of work. IPRally's co-founder and CPO Sakari Arvela breaks down what quality means for Agent, and shares the data behind that claim.

Webinar Q&A follow-up: Your questions about IPRally Agent answered

More people joined our Agent launch sessions than any webinar we've run, and the questions kept coming well after we wrapped. Thank you to everyone who joined! We couldn't address everything live, so we've compiled the most common and insightful ones here and given each the thorough answer it deserves. Whether you attended or are discovering Agent for the first time, we hope this is useful.