Workshops · Topical
AI for Research Collaborations
Half day, about 3 hours, $2,500
Includes: AI Application Assessment ($297 value)
Who it's for
Research office and R&D operations leads asked for the office's position on AI tools, technology transfer directors weighing tools for disclosure triage, corporate R&D and open innovation leads evaluating vendors, and the compliance or IT partner who will approve whatever gets chosen. PIs already using these tools without a policy are welcome. It's for people who buy, approve or govern tools, not people who build them.
The problem it fixes
Research offices on both sides are adopting AI tools one person at a time, usually for drafting and summarizing, with no shared view of which tasks the tools suit or what happens with confidential partner data. The questions that follow (Can the sponsor's unpublished results go into this tool? Who is the inventor if the tool helped?) land on a compliance office with no framework for answering them. This workshop sorts your partnership workflow into tasks where AI helps, where it doesn't, and where a rule is needed, and leaves the team with a selection method and a one-page use policy.
Agenda
Four blocks of about 45 minutes.
- 0:00 to 0:45. Debrief of your AI Application Assessment results. We read the scores for each candidate application and agree which are worth pursuing, which scored low for a reason that won't change, and which depend on a data decision.
- 0:45 to 1:30. Where it helps and where it doesn't. Partner search and literature triage, first drafts of reports, meeting notes and action tracking on one side; judgment calls, trust-building and negotiation on the other. We test each against your own examples.
- 1:30 to 2:15. Selecting a tool. Working from our free AI Tool Selection Guide, a session handout, we set the questions a vendor must answer: where the data goes, whether it trains a model, who sees the outputs, what the contract says about confidentiality.
- 2:15 to 3:00. Governance basics. The four functions of the NIST AI Risk Management Framework as a checklist your compliance office will recognize, the EU AI Act risk tiers if a partner is in the European Union, and the USPTO's revised inventorship guidance of 26 November 2025 for AI-assisted inventions, the reason your disclosure process needs a line on tool use. Helikon Labs is not a law firm and this is not legal advice.
What you bring
- Your AI Application Assessment results
- One or two workflows you're considering a tool for, and any tool already in use
- Whoever approves tools, whoever owns research data policy, and the people who would use the tool
What you leave with
- Your collaboration tasks sorted into suited to AI assistance, not suited, and needing a data rule first
- A vendor question list tailored to your data
- A one-page AI use policy, ready for compliance review
- A note on what to add to your invention disclosure process
Included instrument and pre-work
The AI Application Assessment is issued at booking and completed two weeks before the session. It scores the candidate AI applications across your collaboration workflow, so the first block starts from your results, not a vendor's demonstration. The AI Tool Selection Guide is not pre-work, it's a handout for the third block.
Logistics
Capped at 12 participants: J. Richard Hackman's team research found that coordination and candor erode once a working group climbs into the double digits (Hackman, interviewed in Harvard Business Review, 2009), and a working session needs everyone contributing, so larger teams add breakouts or split into additional sessions. Virtual by default, onsite available with travel billed at cost at federal per-diem rates. Series pricing is 15% off three or more workshops. We schedule within one business day of booking.
Frequently asked questions
Can we run it for a mixed university and company group?
Yes, and it can be even more effective this way. Both sides hear the same debrief in one room, see where their results already differ, and make decisions that would otherwise take months of meetings.
What if we already own the tool?
The instrument's price is not deducted from the workshop fee. We use your existing results as the pre-work, and you keep the second license for a second team or partnership.
Do we need technical staff in the room?
No. The decisions are which tasks to assign to a tool, how to choose one and how to govern it, and those belong to the people who manage your research. An IT or security partner helps with the third block's data-handling questions, but is not required; we write those down.
Not sure it fits? Book a free 30-minute scoping call.
Sources
- NIST, 2023, AI Risk Management Framework (AI RMF 1.0). www.nist.gov/itl/ai-risk-management-framework
- European Union, 2024, Regulation (EU) 2024/1689, the AI Act. eur-lex.europa.eu/eli/reg/2024/1689/oj
- USPTO, 2025, Revised inventorship guidance for AI-assisted inventions, Federal Register. www.govinfo.gov/content/pkg/FR-2025-11-28/pdf/2025-21411.pdf
- Coutu, 2009, Why Teams Don't Work (interview with J. Richard Hackman), Harvard Business Review. hbr.org/2009/05/why-teams-dont-work