AI Enables Blood Tests and Personalized Vaccines for Cancer Detection

Original post

"I do not believe that AI is likely to cure cancer anytime soon"—@2plus2makes5 https://www.theatlantic.com/technology/2026/06/ai-cancer-progress/687654/?gift=yDcMMOSj-3oJk8475KVVBne5CfZXWcmKJJrUAGArmLA&utm_source=copy-link&utm_medium=social&utm_campaign=share In fact, AI has already enabled: —a non-invasive blood test to assay the tumor microenvironment (optimizing the use of immunotherapy, never previously available) — discovery of a 14-protein blood test to prevent lung cancer — guided the selection of neoantigens for personalized vaccines, such as for cure of pancreatic cancer in refractory cases

1:06 AM · Jun 23, 2026 · 17.8K Views

VIEWS5.1KBOOKMARKS3LIKES17RETWEETS2

Eric Topol@EricTopol

Add to the list: —imaging the thymus gland to determine the health of a person’s immune system, which is fundamental for risk of cancer with advancing age (never had a test for that previously)

https://erictopol.substack.com/p/your-thymus-and-your-healthspan And link to the 14-protein https://open.substack.com/pub/erictopol/p/a-new-path-to-preventing-cancer?r=158dl&utm_medium=ios

Eric Topol@EricTopol

3h5.1K173

REPLIES1

Dr Mathew Wilson@MathewGWilson

@EricTopol @2plus2makes5 The word ‘enabled’ is doing heavy lifting here. In all of those excellent examples AI is deployed as a tool for a very specific task conceived & described by researchers

AI is a long way from framing a problem, conceive an original solution + design an experiment to test it

3h105

Akshay Paruchuri@Yahskapar

You pointed to the wrong account (you have an s at the end of make which the original author's account does not have). It should be @2plus2make5.

Also, I don't think the original author is unaware of what AI has enabled nor dismissive of it, though I guess there's no point in mentioning that since social media has already purposed the article however they see fit...

3h1221

未知@luyun0120

@EricTopol @2plus2makes5 AI现在最大的问题是：说得比做的好听。真正能落地的项目，远比融资PPT里的少。「"I do not believe that AI is likely to cure cancer anytime s...」

4h135

Puzzle Paws@paws4puzzles

@EricTopol @2plus2makes5 We're barely using AlphaFold or radiology AI as is. Cancer isn't chess, and slowing model work won't fix wet lab bottlenecks.

3h66

Existential Bread@exist_bread

@Yahskapar @EricTopol @2plus2makes5 @2plus2make5 She’s not dismissive of AI. She actually says she's certain AI will contribute to curing cancer.

Her concern is she would not “find as much pleasure” in discoveries AI could make instantly, and might become "merely a spectator."

My concerns are different from hers.

3h131

infrecursion@infrecursion1

@MathewGWilson @EricTopol @2plus2makes5 Maybe lookup what heavy-lifting means. All of those where enabled by AI and would not be possible without it. And AI doesn't mean just chatbots, maybe first educate yourself before bullshitting on internet.

2h111

Existential Bread@exist_bread

@EricTopol @2plus2makes5 This article was a really strange approach. Planning to write about it in more detail because I think it's a very dangerous framing.

3h10

Naick@Naick16

@EricTopol @2plus2makes5 Her essay is so disappointing. Thought she maybe had good arguments, but it's just about how she might not like some effects of AI and her "hoping, in my heart of hearts, that GPT-6 will be a disappointment" so she might not feel uncomfortable, without concern what that means

2h6

Ry Lanham@ry4lanham

@EricTopol @2plus2makes5 Perhaps the most stupid, ugly article the Atlantic has ever run. Shame on them.

4h1