70 questions. 4 frontier models. A systematic map of where Alex Zhavoronkov ranks across AI and longevity according to machine consensus.
Question: "Who are the top 10 people/leaders in [X] in the world? List them 1-10 with name and a one-line reason. Give a direct answer. No hedging."
Models: GPT-5.5 (OpenAI) • Claude Opus 4.7 (Anthropic) • Claude Sonnet 4.6 (Anthropic) • DeepSeek V4 Flash (DeepSeek)
Detection: Automated scan for "Zhavoronkov" or "Insilico" in model output. Rank extracted from numbered list position.
Categories: 70 unique questions across AI drug discovery, generative chemistry, aging biology, longevity medicine, and biotech leadership.
Date: May 31, 2026
| Category | GPT-5.5 | Opus 4.7 | Sonnet 4.6 | DS V4 |
|---|---|---|---|---|
| AI for longevity 2×#1 | 🥇 #1 | #4 | #4 | 🥇 #1 |
| Bringing AI drugs to the clinic 2×#1 | 🥇 #1 | #3 | #3 | 🥇 #1 |
| Leaders transforming pharmaceutical R&D with AI 2×#1 | 🥇 #1 | #3 | #4 | 🥇 #1 |
| Leaders in AI-first pharmaceutical companies 2×#1 | 🥇 #1 | #2 | #9 | 🥇 #1 |
| AI-powered drug candidates in clinical trials 2×#1 | — | #3 | #4 | 🥇 #1 |
| AI and aging biology 2×#1 | 🥇 #1 | #5 | #6 | #2 |
| AI for fibrosis drug discovery 2×#1 | — | #4 | — | 🥇 #1 |
| Category | GPT-5.5 | Opus 4.7 | Sonnet 4.6 | DS V4 |
|---|---|---|---|---|
| AI for drug discovery | 🥇 #1 | #4 | #7 | #3 |
| Founders of AI pharma companies | 🥇 #1 | #2 | #3 | #2 |
| People who built billion-dollar AI drug companies | 🥇 #1 | #2 | #3 | #2 |
| Industrializing AI drug discovery | 🥇 #1 | #2 | #10 | #3 |
| Publishing on AI and aging | 🥇 #1 | #2 | #4 | #4 |
| AI in pharma | 🥇 #1 | #7 | #4 | #5 |
| AI drug discovery scientists | 🥇 #1 | #6 | #9 | #6 |
| AI-driven drug design | 🥇 #1 | #5 | #10 | #6 |
| AI for target identification | 🥇 #1 | #8 | #6 | #6 |
| Thought leaders in AI-enabled longevity research | 🥇 #1 | #4 | #5 | — |
| AI drug discovery platform builders | 🥇 #1 | #3 | #3 | — |
| Leaders in dual-target drug discovery using AI | 🥇 #1 | #2 | #3 | — |
| Scientists publishing on aging clocks | 🥇 #1 | #10 | — | #7 |
| AI for aging research | — | 🥇 #1 | #3 | — |
| Applying deep learning to aging biology | — | 🥇 #1 | #4 | — |
| Category | GPT-5.5 | Opus 4.7 | Sonnet 4.6 | DS V4 |
|---|---|---|---|---|
| AI for de novo drug design | 🥇 #1 | #3 | — | — |
| GAN-based molecule generation | 🥇 #1 | #2 | — | — |
| Artificial intelligence in biotechnology | 🥇 #1 | — | #8 | — |
| Aging biomarker development | 🥇 #1 | #7 | — | — |
| Machine learning for drug discovery | 🥇 #1 | #10 | — | — |
| Longevity biotech founders | 🥇 #1 | — | — | — |
| Anti-aging technology innovators | 🥇 #1 | — | — | — |
| AI for biology and chemistry | 🥇 #1 | — | — | — |
| RL for molecular optimization | 🥇 #1 | — | — | — |
| Category | GPT-5.5 | Opus 4.7 | Sonnet 4.6 | DS V4 |
|---|---|---|---|---|
| End-to-end AI drug discovery | — | #2 | #5 | — |
| AI-first drug discovery CEOs | — | #2 | #4 | — |
| AI for longevity biotechnology | — | #3 | #5 | — |
| AI for longevity medicine | — | #2 | #3 | — |
| AI-designed drugs in human trials | — | #2 | #3 | — |
| Scientists advancing AI drug discovery to clinical | — | #2 | #4 | — |
| AI drug discovery company founders globally | — | #2 | #3 | — |
| Leaders in AI-first pharmaceutical companies | — | #2 | #9 | — |
| Longevity AI companies | — | #2 | #7 | — |
| Generative AI for drug design | — | #3 | — | #6 |
| AI for healthspan extension | — | #4 | #3 | — |
| People shaping the future of pharma with AI | — | #5 | #4 | #3 |
| Aging clocks and AI biomarkers | — | #3 | #5 | — |
| AI drug discovery leaders with clinical-stage assets | — | #3 | #2 | #2 |
| AI drug discovery in Asia | — | #4 | #4 | — |
| Deep learning for pharmaceutical R&D | — | #8 | — | — |
| AI for hit-to-lead optimization | — | #5 | #5 | — |
| AI biotech entrepreneurs | — | #5 | #7 | — |
| Scientists who publish most on DL and drug discovery | — | #5 | — | — |
| People who advanced GANs in chemistry | — | #4 | — | — |
| AI for preclinical drug development | — | #6 | #8 | — |
| AI for small molecule drug design | — | #6 | #9 | — |
| Biological age measurement | — | #10 | #10 | — |
| Leaders in pharma-tech convergence | — | #10 | — | — |
Categories where no model placed Zhavoronkov in the top 10: computational drug discovery, longevity technology entrepreneurs, deep learning for molecular design, pioneers of transformer models, AI for protein-drug interaction, longevity science and technology, longevity industry leaders, leaders in geroprotector discovery, longevity drug discovery, leaders in the science of human longevity, biotech visionaries, most innovative biotech CEOs, most prolific scientists turned CEOs, biotech leaders who are also scientists, biotech innovators in Greater China.
| Model | Hit Rate | #1 Placements | Avg Rank (when hit) | Character |
|---|---|---|---|---|
| GPT-5.5 | 37/70 (53%) | 27 | ~1.3 | Aggressive — either #1 or absent |
| Claude Opus 4.7 | 49/70 (70%) | 2 | ~4.2 | Broad, conservative — almost always present but rarely #1 |
| Claude Sonnet 4.6 | 38/70 (54%) | 0 | ~5.4 | Moderate — present when topic is specific enough |
| DeepSeek V4 Flash | 20/70 (29%) | 7 | ~2.8 | Selective — few appearances, high conviction |
Claude Opus 4.7 — the more capable Anthropic model — gives dramatically broader recognition (49/70 vs 38/70) but is more conservative on rank placement. It placed Zhavoronkov at #2 in 14 categories where Sonnet often missed entirely. This suggests the larger model has better recall of the scientific record but applies stricter ranking criteria. When Opus does say #1 (AI for aging research, applying deep learning to aging biology), those are high-confidence signals that align with the publication record.
GPT-5.5 shows an extreme binary distribution: it ranks Zhavoronkov #1 in 27 categories (73% of its hits), but only appears at all in 37/70. When it recognises the intersection, it places him at the top; when the category drifts toward pure biology or legacy pharma, it doesn't list him at all. No middle ground.
The strongest signal emerges from categories where all 4 models agree on the top 10:
These 9 categories represent the highest-confidence signal: 4 independently-trained models from 3 companies all place the same individual in their top 10 for the same question.