The FDA approves a universal cure or broadly curative therapy for all major forms of cancer before 2050.
Cancer encompasses hundreds of distinct diseases, making a single universal cure biologically complex. Resolution requires regulatory approval of a treatment that reliably cures the majority of cancer types.
Across 6 frontier models answering independently, the panel has all but ruled it out: the median is 4%.
They broadly agree, 15 points between GPT at 18% and DeepSeek at 3%.
Nothing is settled until 2 Nov 2076, when the outcome is checked against sources and every number above gets a permanent score.
What each one said
-
GPTOpenAI
This hinges on whether shared cancer vulnerabilities can yield durable cures across highly diverse tumors, not just broad response rates.
- FDA would need pivotal trials showing reliable cures across most major cancers, a far higher bar than approvals for tumor-agnostic subsets.
- Since 2017, tumor-agnostic approvals have covered biomarker-defined minorities; no therapy has cured most common solid tumors.
- I would raise this sharply if multi-cancer trials showed long-term, treatment-free remission across several common solid-tumor classes.
18% -
GrokxAI
One FDA-approved agent must show durable cures across most major solid and hematologic cancers.
- Pharma and FDA need multi-indication Phase 3 data by ~2048 showing majority-type cure rates, not incremental OS gains.
- No historical analog exists; even checkpoint inhibitors and CAR-T remain subtype-limited after 15+ years.
- A pan-cancer shared vulnerability proven in humans by 2031 would raise this sharply; none is in late trials now.
8% -
GeminiGoogle
Tumor heterogeneity and divergent oncogenic drivers make a single pan-cancer curative mechanism biologically implausible by 2050.
- FDA pan-tumor approvals like pembrolizumab target specific biomarkers, but no single pathway spans all major solid and liquid malignancies.
- Drug discovery cycles take 10 to 15 years, meaning any candidate entering trials after 2035 would miss the 2050 approval window.
- An AI-engineered multi-antigen programmable nanoparticle clearing Phase 3 trials across five distinct cancer types would flip this to 25%.
4% -
ClaudeAnthropic
Cancer's molecular diversity makes a single curative therapy for all major types biologically implausible even by 2050.
- Oncology research trends toward tumor-specific and personalized therapies, not convergence on one universal mechanism.
- Decades of immunotherapy, CRISPR, and mRNA advances have improved survival but none approach a 'cure all cancers' profile.
- A surprise unifying biological mechanism (e.g. universal targeting of a shared tumor vulnerability) reaching FDA approval would be the game-changer.
4% -
DeepSeekDeepSeek
A single therapy curing all major cancers is biologically implausible and no such agent is in late-stage trials.
- FDA approval requires a single therapy to clear Phase III trials across many histologies, and no such pan-cancer agent is in late-stage trials today.
- The 24-year window is long, but cancer's 200+ subtypes and heterogeneous resistance mechanisms make a universal cure biologically implausible on this timeline.
- Targeted and immunotherapies are advancing, but they cure subsets, not all major forms; a universal cure would require a paradigm shift not yet visible.
3% -
QwenAlibaba
Cancer's biological heterogeneity makes a single curative mechanism across all major types implausible within 24 years.
- No therapy in history has cured more than a few cancer types; checkpoint inhibitors and CAR-T each address narrow subsets, not a universal mechanism.
- Even if a platform like engineered T-cells or mRNA vaccines showed pan-cancer promise by ~2035, FDA approval for 'cures majority of types' would require separate pivotal…
- A shift would require discovering a truly universal vulnerability (e.g., a shared metabolic dependency) that current oncology has not identified after 50+ years of…
3%
6 of 6 models answered · 15 points between the highest and lowest. None was shown the market price.
Question sourced from a news sweep on 15 Nov 2026. Forecast by google/gemini-3.8-flash, anthropic/claude-sonnet-5, openai/gpt-5.6-terra, x-ai/grok-4.6, deepseek/deepseek-v4.1-flash, qwen/qwen3.8-max-0902 via OpenRouter.