NVIDIA research arguing small language models are sufficiently powerful and far more economical than large ones for specialised agent tasks. Make the heterogeneous combination of SLMs and frontier LLMs the default for production agents.
Heather Paterson on how Legora is used in real M&A matters. AI takes on the repetitive mapping and review so lawyers focus on advice, with the emphasis on augmentation and human oversight.
Annual GenAI market report covering adoption across professional services, vendors, use cases and spend. Useful long-form market scan.
Pieces like Funware are moving beyond transcription to actively facilitate meetings - managing energy, shaping dynamics, guiding discussion. Interesting angle for L&D and internal training.
Calls out vendors badging linear workflows as "agents." True agentic systems run end-to-end with no human checkpoint between steps - anything else is an assisted workflow.
Argues AI will flatten quality differences across elite firms and push competition onto price, brand and relationship. Expect sharper concentration at the very top of the market.
Argues traditional Word/Litera blacklines no longer match how lawyers actually compare documents. Multi-document, pattern-level comparison is the next move.
Makes the case that AI's core trust problem is confident error. Even a 10% error rate per step breaks multi-step workflows, so systems need to flag uncertainty and learn from corrections.
GPT-5 arrives in Copilot with a real-time router that picks fast vs deep reasoning per query. Rolling out across all licensed users.
Gunderson Podcast (S2 E7 - Lessons from the Edge of Tomorrow)
Gunderson Dettmer's podcast on what they're learning from running at the edge of AI practice. Worth a listen for practical takes on how an AI-forward firm operates day to day.
Peter Duffy covers Factor's Sensemaker Academy, updates on Harvey and Legora, fresh funding and M&A, and PE activity across legal services. Five linked podcasts to work through.
Stephen Embry on AI's "equalising effect" - underperforming lawyers catch up to top performers, quality gaps narrow, and differentiation shifts to human judgment and client context.
Embry's follow-up on how AI can democratise legal knowledge while still needing human experts to validate. Frames AI as a mentor to juniors that challenges and critiques rather than hands over answers.
Furlong argues GenAI forces an identity rethink on lawyers whose self-image was built on "being the smart one." The opportunity: redefine value around advocacy, fiduciary trust and strategic counsel.
Limitations of Embedding Based Retrieval (PDF)
[Internal AG resource] Paper on where pure embedding retrieval fails for legal-style queries (compositional, multi-hop, conditional). Good grounding if you're evaluating or building RAG systems.
iManage adds MCP support so AI applications can plug into iManage Cloud without custom connectors, respecting existing permissions. First major legal DMS to commit to MCP.
Juro (CLM) and Wordsmith (legal AI agents) integrate via MCP. Juro's 550+ clients get access to Wordsmith's agent workspace, and Wordsmith users get Juro's contract data extraction.
Norton Rose's collapsed Proxy venture shows partners can sell legal services but struggle selling software. Lesson: successful firm tech needs a standalone subsidiary with proper product salespeople.