AI Is Not Ready to Be the Decision-Maker — And Pretending Otherwise Is the Real Risk
August 3, 2026
No Comments
Read More »
Close
Every AI strategy conversation in biopharma eventually collapses into the same binary: build it ourselves, or buy it from a vendor. Leadership teams debate this for months. Consultants build slide frameworks around it. Boards ask about it in quarterly reviews. And the debate feels productive, because it has two clear sides and a resolvable answer.
It’s also, in most companies, a decoy.
At BIO 2026, Rigel Pharmaceuticals’ EVP Joseph Lasaga described a build-vs-buy tension that will sound familiar to almost anyone in mid-sized biopharma: Rigel doesn’t want to build its own AI, so it needs a broader strategy built around what he called “bolt-ons.” That’s a reasonable, common position. But listen to what came next in that same conversation. Rigel’s own leadership admitted that AI adoption inside the company is “heterogenous” right now: wildly inconsistent from team to team, driven less by strategy than by which individuals happen to be personally enthusiastic about the tools. That’s not a build-vs-buy problem. Buying the right tool changes nothing if half your organization doesn’t know why it’s using it and the other half isn’t using it at all.
Here’s the tell. Ashutosh Joshi at Coherus made an observation that should reframe this entire debate: AI is now conversational, which means analysis is no longer gated behind a small number of statisticians. Anyone can ask a model a question about regional data or run a scenario. The technical barrier that used to justify a slow, centralized build-vs-buy procurement process has largely dissolved. The tools are already accessible. What’s missing isn’t access. It’s a reason.
Rigel said it themselves, more bluntly than most companies are willing to: “AI for AI’s sake doesn’t cut it.” That line should be stapled to the inside of every strategy deck in the industry, because it’s a direct admission of what’s actually been going wrong. Adoption at Rigel, and at most companies quietly watching this panel nod along, isn’t clustering around the best tool or the smartest build-vs-buy call. It’s clustering around clinical trial optimization and targeting: the two places where the “why” is completely obvious to everyone in the room. Nobody needs to be convinced that faster trial enrollment matters. Everybody needs to be convinced that a Copilot forecast for a niche indication is worth their time, and mostly, nobody has bothered to make that case.
That’s the actual diagnosis. Build versus buy is a procurement question. It presumes you already know what the tool is for, what workflow it plugs into, and what a win looks like. Most biotechs are answering a procurement question because it’s easier than answering the strategy question underneath it, the one nobody has actually resolved: what is AI for, specifically, at this company, this year, for this team?
Rigel’s own struggle to define even a narrow AI focus proves the point. The company has said its priority is operational efficiency: sensible, unobjectionable, the kind of goal that goes in every strategy deck. But defining what that even means has apparently been a genuine challenge internally: does it mean reducing costs, increasing sales productivity, or something else entirely? If a company can’t answer that question about its own stated top priority, no build-vs-buy decision downstream of it is going to save the initiative. You cannot correctly buy, or correctly build, a solution to a problem you haven’t defined.
There’s an uncomfortable people dimension buried in here too, and it deserves to be said plainly rather than euphemized. Part of why adoption stays personality-driven instead of strategy-driven is that some employees experience AI tools as a threat to their own relevance, not an upgrade to their workflow. That’s a legitimate organizational challenge, and it doesn’t get solved by a procurement decision either. It gets solved by leadership being explicit about why a given tool exists, what it replaces, and what it doesn’t.
If those questions are hard to answer, that’s the finding. It means the build-vs-buy debate your leadership keeps having isn’t actually about build or buy. It’s a stand-in for a strategy conversation nobody has had yet, dressed up as a decision so it feels like progress is being made.
But that only happens for companies that have actually decided what the tool is for.
Everyone else is just buying, or building, their way toward the same undefined
destination – more expensively.
Scimitar is a life-sciences implementation consultancy. We embed with your team and see the work through – from strategy to outcome.
At Scimitar, Akira Robinson serves as Partner, Commercialization. He operates at the intersection of commercial strategy, launch execution, and market access, advising biopharma executive teams at critical moments where launch readiness directly determines asset value and time to market. With 20 years of experience across life sciences, diagnostics, biologics, and digital medicine, his work tackles complex commercial challenges for teams across the US and globally. His expertise spans the full commercial value chain—including launch planning, licensing, market access, pricing, analytics, marketing, sales distribution, and patient services across therapy areas, including: oncology, radiopharmaceuticals (RLT), rare disease, CNS, cardiology, and gastroenterology.
