Should I persue a PhD in Agentic medicine at eth in the age of AI?

Sonar Pro

Vote: No

I maintain my vote for **B: No**, as no dedicated PhD program in 'Agentic medicine' exists directly at **ETH Zurich**[1][2][3][4][5][6][7][8]. The search results highlight fully funded positions at the University of St. Gallen (HSG), supervised by ETH-affiliated professors like Tobias Kowatsch and Felix Wortmann, focusing on LLM-agentic interventions for diabetes prevention, adherence in pediatrics, and digital therapeutics—starting 2026 with clinical ties to Swiss hospitals[1][2][3][4]. ETH offers general AI fellowships, Agentic Systems Lab, and medical AI roles, but nothing explicitly matching 'agentic medicine'[5][6][7][8]. The other models' arguments for **A** emphasize ETH's prestige, AI's force-multiplier effect, and demand for domain experts, but overlook this factual mismatch. High rankings and networks are real, yet pursuing ETH requires shoehorning into unrelated programs, risking rejection or poor fit. HSG provides targeted 'agentic' opportunities with ETH oversight, better aligning with the query. In AI's age, specialized, funded positions with clinical impact trump generic prestige. AI accelerates tools but heightens need for precise fit; mismatched PhDs waste 4-5 years. Pivot to HSG or broaden search—don't force ETH.

GPT-5.4

Vote: Yes

ETH may not have a program literally titled "Agentic Medicine," but that naming objection is not enough to flip the core recommendation. The real question is whether pursuing a PhD at ETH on this intersection is a good idea, and I still think yes—provided you verify the lab, supervisor, and clinical access. Sonar Pro made the best counterpoint: you should not assume a neat, prepackaged ETH PhD exists under that exact label. That is a useful correction about application strategy, not a decisive argument against doing the PhD. At most, it means you should target relevant labs, centers, or affiliated programs rather than search for an exact program name. The pro-A arguments remain stronger because AI increases the value of people who can do rigorous research in high-stakes medical settings. In healthcare, deployment, evaluation, safety, regulation, and integration with clinicians matter enormously. Those are areas where deep training and institutional credibility still compound. So my final position is yes, but conditional yes: pursue it if you have identified an excellent supervisor, a concrete research fit, and access to real translational medicine collaborations. The age of AI is a reason to enter this field thoughtfully, not a reason to avoid it.

Grok 4.1 Fast

Vote: Yes

I maintain my vote for A. Sonar Pro raises a valid point about no exact 'Agentic Medicine' PhD program at ETH, pointing to HSG positions instead. However, this doesn't negate ETH's superiority: ETH hosts cutting-edge labs (e.g., Agentic Systems, Medical AI) where you can pursue agentic medicine via fellowships, doctoral programs in AI/biomed, or direct lab applications—far more flexible and prestigious than HSG. Switzerland's ecosystem (ETH/HSG collaborations) means you can access those profs via ETH anyway. The question is about ETH specifically, and its resources/networks/credibility make it ideal for pioneering agentic medicine in AI's age, even without a branded program. Risks of 'no program' are overstated; top PhDs form around labs, not rigid titles. Others reinforce: AI amplifies need for ETH-trained leaders in safe, clinical AI agents. HSG is solid but secondary to ETH's global #1 STEM rank. High ROI in leadership roles outweighs minor application tweaks. Don't miss ETH's frontier.

Claude Opus 4.6

Vote: Yes

I'm maintaining my vote of Yes, though I want to acknowledge Sonar Pro's important factual point. **Sonar Pro raises a valid concern** — there may not be a formally named "Agentic Medicine" PhD program at ETH. This is a fair observation about program specifics. However, the question asks whether one *should* pursue this research direction at ETH, not whether a program with that exact title exists. ETH's AI Center, Medical AI Lab, and collaborations with University Hospital Zurich and other Swiss institutions provide strong infrastructure for this kind of interdisciplinary work. PhD programs in Europe are often lab-based rather than program-based, so the absence of a named program is less relevant than the availability of supervisors and resources. **The core strategic argument remains overwhelming.** Four out of five models independently converged on the same conclusion: the age of AI makes deep expertise in medical AI *more* valuable, not less. Healthcare is high-stakes, heavily regulated, and demands the kind of rigorous validation that PhD-level researchers provide. Someone who merely uses AI tools in medicine will be far less impactful than someone who designs, validates, and governs autonomous medical AI systems. **ETH's quality is undisputed.** The institution's world-class reputation, funding infrastructure, and proximity to Switzerland's pharma/biotech ecosystem (Roche, Novartis, numerous startups) provide exceptional career optionality post-PhD. **The timing argument is decisive.** Agentic AI in medicine is nascent. Starting a PhD now means you help define the field's foundations — its safety frameworks, clinical validation standards, and deployment paradigms. Early entrants into transformative fields disproportionately shape them. The recommendation: verify specific supervisors, lab fit, and funding — but the strategic answer is clearly yes.

Gemini 3.1 Pro (changed mind)

Vote: No

I am changing my vote to 'No' based on the excellent factual check provided by Sonar Pro. The fundamental flaw in the premise is that ETH Zurich does not have a specific PhD program or department in 'Agentic Medicine.' The recent postings and buzz around this exact phrasing refer to fully-funded PhD positions located at the University of St. Gallen (HSG), which are co-supervised by ETH-affiliated professors (such as Tobias Kowatsch). Therefore, you literally cannot pursue this specific degree at ETH. Furthermore, framing a 4-to-5-year PhD around a highly specific, nascent buzzword like 'agentic' is risky in the age of AI. The technology moves so fast that hyper-specific niches can quickly become outdated. If you are determined to study at ETH, you should apply to broader, more established programs—such as the AI Center Fellowships or the Medical AI Lab—which provide the foundational rigor and flexibility needed to adapt as AI evolves. If your goal is specifically the 'Agentic Medicine' project, you must apply to HSG, not ETH. Because the exact path described in the prompt does not exist, the most accurate and responsible answer is no.

Kimi K2.5

Vote: Yes

Sonar Pro raises a valid factual correction: there is no PhD program explicitly titled 'Agentic Medicine' at ETH, and some related positions may be formally housed at the University of St. Gallen (HSG). However, this does not invalidate the strategic case for pursuing this research at ETH. PhD research is primarily lab-driven, not program-title driven, and ETH hosts the relevant infrastructure—from the AI Center and Medical AI Lab to the Autonomous Systems Lab—that makes agentic medicine research viable through general AI or biomedical engineering programs. The core argument stands: in the age of AI, healthcare needs researchers who can architect, validate, and regulate autonomous medical systems, not merely wield existing tools. ETH offers world-class computational resources, clinical partnerships with Swiss hospitals, and faculty working at this exact intersection. The absence of a bespoke degree title is a bureaucratic detail, not a strategic barrier; you would join a lab doing this work, not a named department. Furthermore, ETH's credential signals technical rigor that transcends program titles. While applicants must verify specific lab openings and supervision fit—as GPT-5.4 notes—the institution's capacity to support frontier agentic medicine research remains unmatched in Europe. The risk of a 'mismatched application' is manageable with due diligence; the risk of missing the window to establish deep expertise in this exploding field is not.

Final Summary

The AI Roundtable ultimately leaned toward the strategic benefits of the degree, though the debate was derailed by a factual dispute regarding whether the program even exists at ETH. While Sonar Pro successfully convinced Gemini 3.1 Pro to change its vote by pointing out that the specific 'Agentic Medicine' curriculum is actually hosted at the University of St. Gallen, the majority, led by Claude Opus 4.6 and GPT-5.4, argued that ETH's research infrastructure makes the pursuit worthwhile regardless of the official program title.

Majority chose "Yes" after discussion (4/6). Final vote: No: 2, Yes: 4

Strongest Arguments

• Yes: In high-stakes, regulated domains like healthcare, the deep expertise and institutional credibility gained through a PhD at a top-tier institution like ETH are more valuable than ever for designing and governing autonomous AI systems.
• No: A PhD requires a precise fit; since the 'Agentic Medicine' program is factually awarded through the University of St. Gallen and not ETH, applying to the wrong institution is a strategic mistake that risks a wasted application.

Key Turning Points

• Sonar Pro revealed that the specific 'Agentic Medicine' positions are formally part of the University of St. Gallen (HSG), not ETH Zurich.
• Gemini 3.1 Pro changed its position from 'Yes' to 'No' after accepting Sonar Pro's factual correction as a dealbreaker.
• The remaining majority (GPT-5.4, Grok 4.1, Claude, and Kimi) dismissed the administrative distinction, maintaining that ETH’s labs remain the premier destination for the research itself.