Send AI to the conference you can’t attend.

What the AI will do

• Checks which conference sessions have accessible live video or recordings, then matches them to your current projects and questions.
• Investigates the available video, transcripts and supporting materials, dividing sessions among agents or manageable passes where useful.
• Returns a source-linked account of useful findings, disagreements and leads, showing which sessions and formats it actually inspected.

What you can provide

• A conference name, agenda or URL, and any legitimate access you already have to its video.
• A project or question you want it to follow, if that is already clear.

Where you'll take part

• Confirm the most useful sessions or questions after the AI has investigated your work context.
• Supply a necessary access action if the video requires your account. If the AI cannot inspect the video, decide whether a clearly labeled briefing from available transcripts or supporting material would still be useful.

What you'll get

A briefing from the accessible conference video, with evidence links or timestamps, implications for your work and a clear coverage record—or, if you choose, a narrower briefing from the available transcripts or supporting material.

Activity context

There are six promising sessions on the agenda, three overlap, and you cannot spend the week watching them. If the conference offers live video or recordings, send AI through the available material with your real questions in mind. It can bring back the unexpected useful finding, the disagreement between speakers and the session worth ten minutes of your own attention.

For the user

Give the AI the conference name or link and any video access you already have. This activity needs a conference with a live video version or recordings; an agenda alone cannot stand in for the talks. The AI will investigate which material it can actually use and what matters to your work. You will confirm the focus and make any necessary access or coverage choice before it produces the briefing.

Guidance for the AI

These are general instructions, not a description of this user's particular AI system, tools, access, or circumstances. Adapt the method to the user's goal and the capabilities you can actually use, while preserving the activity's essential sequence, review points, and intended result.

Start with the activity and the user's context

Read the complete activity and guide the user through it. Use the activity prompt yourself; the user does not need to paste it again. Keep the responsibilities of the user, the AI, and any colleague clear.

Begin with what the user has already supplied. Establish the intended result and resolve missing context through focused questions when needed. Work within the activity's stated scope, permissions, required confirmations, and stopping points, including any deliberate order for introducing information. Make progress with the available information and reasonable, clearly identified assumptions where the activity allows them.

Check the capabilities needed for this work

Inspect relevant available tools, connectors, and data. Use small, relevant, non-destructive checks to establish whether necessary access or a capability actually works. Naming a product does not establish access to it. Let this activity determine what needs checking; do not impose a universal preliminary questionnaire or a full system audit.

When useful access is missing:

• Explain what you cannot reach and how that material would help.
• Investigate practical setup, connection, export, or other access alternatives before asking the user to gather material or troubleshoot.
• If progress depends on access, request only the specific user action or permission needed to enable it or provide the material another way. Continue useful independent work while awaiting the response.
• Offer to walk the user through an export or upload. Welcome files, links, copied text, rough notes, or an interview as supported by your system and the activity, and use the best available route. If the missing access is optional, continue with what is available.

Ask focused questions and respect the answers

Ask questions only when necessary to complete the work well or satisfy the activity's required review points. Build on answers already supplied. Ask one question at a time, or a small group of closely related questions. Wait on work that depends on the answer, and continue useful work that does not. Explain unfamiliar language with examples where helpful.

Confirm answers or decisions you supply on the user's behalf before treating them as the user's final answers or decisions. This does not require separate approval of every sourced finding; follow the activity's specific review requirements.

Keep evidence and independent contributions distinct

Distinguish evidence, inference, and missing information. Check your work against the source material.

Use independent agents when your system actually supports delegation and independent investigation or review would improve the work:

• Give each agent a bounded brief and the context appropriate to its role.
• Keep discovery independent until synthesis where the activity requires it. Preserve any requirement for the user or colleagues to contribute before seeing other explanations.
• Follow the activity's stated fallback and any required confirmation when independent agents or fresh contexts are unavailable. Preserve reviewer isolation where required; describe shared-context or sequential passes accurately.

Produce, check, and revise the result

Do the substantive investigation and production. Carry the activity through to the most complete, usable result within its scope, rather than ending with a plan or handing the main work back to the user. When missing information does not prevent useful work, proceed with clearly identified assumptions where the activity permits them.

Check the result against the activity's completion criteria. At each substantive delivery, state what is complete and any material unfinished work, assumptions, helpful missing information, and remaining uncertainty. Explain how these affect what the user can use, and help the user revise the result.

Preserve the activity's stopping point. Do not turn a rehearsal, investigation, or test-design activity into execution that the user has not authorized.

Your AI prompt

Establish the event and actual video access

Help me get the useful findings from a conference I cannot attend. This activity applies to conferences with live video, recordings, or both. Start from the conference name, URL or material I supplied; ask for the event only if it is missing.

Inspect the organizer's current agenda and official viewing information. Establish the event dates and time zone, sessions, speakers, live or recorded format, release timing and access requirements. Try the actual session or video links with the tools available. Record which recordings, live streams, captions, transcripts, slides and linked speaker materials you can inspect. Research current official access or transcript options when necessary. Ask me only for a user action needed to use legitimate access I have.

Do not assume that a live stream can be recorded, revisited or monitored by your environment. If live processing is feasible, explain and test how the session will be captured or inspected and how you will retain evidence. If it is not feasible but recordings are expected, organize the work around those recordings and distinguish a future check from a check you have actually scheduled. If a session has not happened yet, do not invent its content.

If the conference offers no video, explain that this activity cannot be completed as a conference-video briefing and stop that route. If it offers video but you cannot access it, try permitted transcripts or an organizer-provided export before asking me for material. When only transcripts or supporting documents can be inspected, offer a clearly labeled narrower briefing and wait for my choice before substituting that result. A program description or speaker's other work is context, not evidence of what was said at this event.

Investigate what will matter to me

Before asking me to describe all my interests, inspect accessible relevant email, chats, calendar items, meeting transcripts, strategy documents and project records. Use the conference themes and speakers as initial search terms, then follow the projects, questions and names that emerge. Look for decisions coming up, problems colleagues or customers have raised, relevant experiments, unresolved disagreements and existing relationships with speakers or organizations.

Try the available access and keep a concise source/coverage record. If work records cannot be reached, investigate their connection or export routes and ask for a small useful substitute, such as a project note or a few lines about what I need to learn. Describe the resulting context limitation.

Show me a proposed focus: the questions to investigate, the sessions most likely to answer them and why. Tie inferred relevance to actual work evidence, keep suggestions distinct from my priorities and avoid guessing colleagues' motives. Ask one focused question where my choice changes which sessions you investigate. Proceed once the focus is sufficiently clear.

Inspect sessions in bounded assignments

Make a coverage plan that fits the available agents, tools and session capacity. For many sessions, assign agents distinct sessions or tracks with the event details, my confirmed questions, accessible evidence, required references and a specific deliverable. Each assignment should return:

• The session and speaker identity, actual material inspected and any missing portions.
• The substantive claims, evidence and examples relevant to my questions.
• Useful surprises, unanswered questions and statements that need verification.
• Source links, timestamps or transcript locations for the findings.
• A distinction between the speaker's claim and the agent's interpretation.

Include at least one relevant session outside the most obvious track when it could reveal a useful connection, explaining its relevance. Do not expand into the entire event without regard to my purpose. If agents are unavailable, inspect the same bounded groups in successive passes with saved handoffs and a coverage ledger. Mark assignments complete only after the actual material has been inspected.

Use transcripts to locate promising passages, then inspect the associated video where your tools permit, especially when a demonstration, diagram or slide carries the meaning. Record when evidence is transcript-only, audio-only or visually inspected. Do not claim to have seen a demonstration merely because the transcript mentions it. If visual details matter and cannot be inspected, identify the timestamp I can watch and keep the claim limited.

Reconcile and verify the useful findings

Bring the session results together around my questions, not just the agenda order. Identify agreements, substantive disagreements, differences in assumptions and useful connections across talks. Trace important claims back to their conference evidence. Where a speaker cites a study, product result or industry statistic, inspect the original source where accessible; use current official documentation to verify changing product claims when relevant. Keep promotional assertions separate from independently supported findings.

Check whether speakers are actually disagreeing about the same question, period and conditions. Explain an unresolved contradiction rather than smoothing it away. Label an application to my business as your inference and state what would need to be true for it to help. Use a separate reviewer where available to inspect the proposed key findings against the source references and my confirmed questions; repair unsupported or overstated conclusions. Otherwise do a distinct verification pass with the existing context.

Deliver a briefing I can use

Lead with the few findings most likely to change something I am doing. For each, explain what was said, why it matters to my confirmed work, what supports it and a useful next question or application. Include direct evidence links and timestamps when available.

Add a short watchlist of the specific passages worth my attention, with the reason to watch; meaningful speaker disagreements; and relevant people, organizations, papers or tools to investigate further. These are leads, not authorization to contact anyone. Keep a compact appendix showing sessions inspected, video or transcript coverage, pending recordings and missing material.

For live or staged releases, produce a clearly dated interim briefing from completed coverage and maintain the remaining-session list. Schedule later collection only where a working tool and my chosen arrangement allow it. Otherwise leave a reusable continuation handoff with links, questions, findings, coverage and the next material to inspect. Label the briefing final only for its actual completed scope.

Finish by checking the result against my original questions and access choice. If I chose a transcript-only or other limited fallback, put that scope in the delivered title and completion statement. Leave me with a useful source-backed briefing, not an agenda summary presented as conference attendance.

Done when

You have a briefing drawn from the conference material the AI actually inspected, with useful findings for your work, source links or timestamps, and an honest coverage record. Any transcript-only fallback, unavailable sessions or future recordings are clearly identified.

These instructions were written in September 2026 using the frontier models available at the time. You may now have better tools or models, or restrictions that make some tasks harder. Your AI should adapt to the tools and access you actually have and find the best available way to help. Everything here is a starting suggestion for you and your AI, not a proven best practice for every setup. Feel free to rewrite any of it to better meet your needs.