Profile avatar
vikas-chowdhry.bsky.social
Bridging the gap between Healthcare & AI. Currently building https://TraumaCare.AI. Chicago Booth, ex-Epic, ex-Parkland Health. Proud immigrant.
43 posts 58 followers 423 following
Prolific Poster
Conversation Starter

Today’s cover of the Murdoch-owned New York Post

Let me tell you about a company that built a new state of matter for Quantum Computing but could not fix this.

They don't want to teach history so they can repeat it.

If you have zero education, but learn how to ask AI models the right questions , in many jobs you will be able to outperform someone with an advanced degree, but who is unwilling to use Large Language Models. Just takes a smartphone, curiosity to experiment and a mindset to learn.

I wrote a book The Creative Destruction of Medicine (2013, title adopted after Schumpeter). This title of the book for what we are now seeing in the U.S. is just The Destruction of Medicine.

One of our supporters shared this with us—people are getting creative tonight!

Delighted to have joined my good friend & colleague Noam Shazeer on a podcast with @dwarkesh.bsky.social for a 2+ hour discussion on early Google, ML hardware, training 1T+ token LLMs in '07, model sparsity, continual learning, and more. Thanks, Noam and Dwarkesh! 🙏 youtu.be/v0gjI__RyCY?...

This is the most relevant article to NIH and research cuts I’ve seen. Imagine if this was today , how many people would be saying “Why are we studying Gila Monsters and their impact on diabetes ? That’s wasted money !” globalnews.ca/news/9793403...

www.wsj.com/business/ent...

🧪 A global survey of 708 radiologists finds burnout is still a major issue, despite AI’s promise to help. 57% don’t use AI regularly, though adoption is highest in Australia. 95% see AI’s limits—radiologists remain essential. 🩺💻

Excited to be working with @drjessilyn.bsky.social to review the most interesting papers from January 2024- January 2025 in AI for healthcare. Share with me your favorites by commenting!

Prior Authorization!

Without Mel Brooks hiring Lynch to make THE ELEPHANT MAN — on the strength of ERASERHEAD alone — he is very likely just a guy who made a weird film at AFI. That whole filmography you love doesn’t happen. We all need sponsors and mentors and patrons. We need people who believe in us.

Fed up with Meta? Avoiding Instagram or Facebook isn’t enough to stop Meta from harvesting and profiting from your private information. Here’s how to limit Meta’s ability to monetize your personal data.

Whatever direct air capture is meant to be, or could be, this is unambiguously what it is becoming: a fossil fuel industry run factory creating single units of greenwashing to sell to the worst emitters maritime-executive.com/article/nyk-...

"Altogether our results suggest that the excitement around the empirical uses of LLMs is warranted, provided researchers guard against training leakage.." Fantastic work by Jens Ludwig @sendhil.bsky.social @asheshrambachan.bsky.social www.nber.org/papers/w33344

Admittedly I went in with low expectations but kinda blown away by integration of Gemini Pro with Google Apps. Way better than what Microsoft has been able to deliver with their O365 apps integration with OpenAI!

A Jarrard survey finds 56% of US nurses distrust AI in admin tasks, 59% in clinical support, and 20% fear job loss. In contrast, 65% of physicians see AI as beneficial (AMA data). 🩺💻 #MLSky

Contrary to popular belief, there is actually a silver bullet to getting value from AI in healthcare. In fact, there are three silver bullets - 1. Workflow 2. Workflow 3. Workflow. :). I elaborate in my @histalk.bsky.social article here: histalk2.com/2024/12/16/r... What do you think?

Canceled my @Office365 Copilot subscription today after a year of getting almost zero value from it. While other AI products are shipping fast & furious and adding value to their subscriptions, Microsoft seems to have stalled after that early burst of energy.

wapo.st/4fd17pA

New LLM model release: Pleias 1.0, "trained exclusively on open data, meaning data that are either non-copyrighted or are published under a permissible license" This is a really big deal: we finally get to try a model that wasn't trained on unlicensed scraped data! simonwillison.net/2024/Dec/5/p...

🧪 Health systems are establishing AI centers to enhance care. Initiatives include Mount Sinai’s Center for AI and Human Health, BJC-WashU’s Center for Health AI, and Hartford HealthCare’s AI Innovation in HealthCare. 🩺🖥️ #MLSky

You probably aren’t being ambitious enough in your AI experiments. Start with completely out there applications, assess them, and then reign in your ambitions from there. You might be inspired or surprised about how good AI is, or you might discover useful limitations earlier.

www.kxan.com/news/texas/t...

If you’re one of the millions of users who are new to Bluesky, you may find this blog useful. 🦋

Mount Sinai unveils a $100M AI and genomics hub in Manhattan, advancing imaging, EHRs, and personalized medicine. The 65,000 sq. ft. center hosts 40 investigators and 250 staff across 12 floors. 🩺🖥️🧬

Of all the AI product subscriptions, I pay the most for Microsoft #Copilot and use it the least since it is pretty terrible. And this despite Copilot have access to all my enterprise data including emails. (I pay for ChatGPT and Perplexity as well).

We often worry about the wrong things. People killed annually by Sharks 10 Elephants 100 Hippos 500 Snails 10,000 Dogs 25,000 Snakes 50,000 People 475,000 Mosquitoes 725,000 🧪

Gorgeous day in Dallas (well, technically Las Colinas).

What’s the CPT code for a platter of crudités?