Good morning {{first_name| Artificial intelligence lovers}}. Artificial intelligence just beat two emergency room doctors with real patient cases in a new Harvard study. Model? o1 preview of OpenAI, which will be released in… 2024.
Millions of users already turn to ChatGPT for health advice every day, but data shows that AI models (preferably not several generations behind) may be ready for a more formal seat in the exam room alongside the doctor, too.
-
Old AI model tops doctors in emergency experience
-
Roundtable Rundown: Our AI Use Cases
-
Create conversion landing pages in Cloud
-
The Pentagon announces new partners in the field of artificial intelligence
-
4 new tools for AI, community workflows, and more
Obinay
Image source: Images 2.0 / Rundown
Rundown: Harvard study published In science, OpenAI’s O1 preview (released in 2024) has just been put through 76 real-world emergencies, where the AI diagnoses patients more accurately than two doctors, despite using only raw EHR text.
-
The study compared OpenAI’s o1 sampling model with two treating physicians across 76 real-world emergencies and three stages of patient care decision-making.
-
In initial emergency triage, the model gave the correct diagnosis 67.1% of the time, compared to 55.3% and 50.0% for the two doctors.
-
The two separate medical reviewers tasked with recording the results were unable to determine which diagnoses came from the model and which came from humans.
-
In one case, artificial intelligence It has been marked A rare infection that eats away at the flesh of an organ transplant patient approximately 12 to 24 hours before the treating physician detects it.
Why it matters: Millions of people already use AI every day to ask health questions, but studies like this show that benefit can also flow in the other direction to doctors themselves. If the model generations behind them are indeed outperforming emergency physicians, imagine what the boundaries within the patient care process could look like.
Along with decoding
Rundown: Unwrap’s customer intelligence platform collects all your feedback – surveys, reviews, support tickets, and social comments – into one view, then uses AI to surface the most actionable insights for delivery to your inbox. Teams at Perplexity, Stripe, and DoorDash rely on Unwrap to ensure the voice of the customer doesn’t get lost.
-
All customer reviews are automatically rated
-
Ask for feedback using Unwrap Assistant, or in your favorite tools using Unwrap’s MCP
-
Real-time alerts of reactions as they appear
-
A clear view of customers’ feelings
RUNDOWN ROUND TABLE
Image source: Ideogram/The Rundown
Rundown: The Rundown Roundtable is a weekly feature in which we poll The Rundown staff members about how we use AI in our work and daily lives.
Jennifer, technology and robotics writer: Last year, my daughter was diagnosed with a rare autoimmune disease. Like any parent in this situation, I wanted to understand everything I could so we could make the best decisions for her care. I used both Gemini and ChatGPT to help me sort through the medical literature, treatment options, potential side effects, and how other countries handle treatment.
I also turned to ChatGPT to learn about the country’s leading specialists – and fortunately, our referring doctor turned out to be one of them, which gave us further reassurance. More than anything, it has helped us demystify the disease and make us feel confident that we are doing everything we can. She’s doing really well when she’s on her medication, and seeing her recover has been a huge relief.
Shubham, Editor: I use ChatGPT as a label-reading filter for packaged snacks, uploading product images and asking it to flag hidden sugars and oils like palm and preservatives, then comparing the options against a strict checklist of clean, minimally processed ingredients and decent macros.
It’s especially useful for deciphering lesser-known terms and heavy jargon on the back label, and translating things like INS numbers, stabilizers, and emulsifiers into plain English so it’s clear what’s actually being consumed. Instead of trusting claims on the front of the package like “multigrain” or “sugar-free,” they break down what’s inside, show the trade-offs, and narrow down the choices to the best options available online.
Artificial intelligence training
Rundown: In this guide, you’ll learn how to use Anthropic’s new AI design tool – Claude Design – to create four high-converting landing page mockups for your website.
-
Go to claude.ai/design, select the wireframe, click Create, and describe who the page is for, the product, and the action visitors should take. Do not press “Send”
-
Take a screenshot of a landing page you like (search the top pages in your niche), as well as a page that does millions of daily transactions, such as Amazon or eBay
-
Press Submit with summary and screenshots. Ask Claude to give four different variations of the mockup. Answer any follow-up questions and wait 2 to 5 minutes
-
Refine with comments. Click on any item and leave a note like “Rewrite CTA to be results-specific” or “Add testimonial here.” Claude implements the change
Pro Tip: Click Share > Deliver to Claude Code > Submit to Claude Code Web to get Claude Code to create and publish the final website for you.
Provided by SCRUNCH
Rundown: Scrunch is an AI customer experience platform that optimizes your site for AI bots – the new VIPs that decide whether your brand gets named, cited, or skipped when someone asks ChatGPT, Perplexity, or Claude about your category.
-
Learn how AI reads your site today
-
Discover barriers and gaps in content
-
Deliver optimized pages directly to AI agents
-
Appears in More AI Answers
Artificial Intelligence and the Pentagon
Image source: Images 2.0 / Rundown
Rundown: Pentagon He added 8 AI companies into their secret networks while excluding Anthropic, even as the Washington Post reports that new contracts have the same limits as autonomous weapons and surveillance Anthropic has been blacklisted.
-
The names of the official agreement list are SpaceX, OpenAI, Google, Nvidia, Reflection, Microsoft, AWS, and Oracle, adding companies to the classified networks.
-
The War Department said the new deals “will accelerate the shift toward establishing the U.S. Army as an AI-first fighting force.”
-
Department of Defense CTO Emil Michael He said CNBC said Anthropic’s supply chain risk label remains, but described the Mythos model as a “discrete national security moment.”
-
Anthropic’s exclusion comes days after the White House opposed the broader Mythos rollout over computing concerns affecting its reach.
Why it matters: It seems the White House wants to have its cake and eat it too – and they both continue to avoid Anthropics while also wanting priority access to their Mythos model despite the blacklist. There are also some interesting names on that list, namely Reflection, which has raised $2 billion from 1789 Capital, a fund backed by Donald Trump Jr.
-
🗣️ Custom sounds – Reproduction of sounds with short clips for use in Grok applications
-
🦮 manuscript pets – OAI mobile companions to track the active work of Codex
-
🎶 Eleven music – A platform for AI song generation, remixes, and creator payments
OpenAI It has been shipped Codex Pets, animated desktop companions that let you track an agent’s progress without returning to the app.
OpenAI CEO Sam Altman Announce That OpenClaw users can use their ChatGPT subscriptions within the proxy tool, taking a stand against Anthropic’s Restrictions.
Maryland I fell The first US ban on AI-based grocery pricing, with fines of up to $25,000 for stores caught using personal shopper data to raise prices.
sag-avtra Believer New AI guardrails in the studio’s four-year deal, with the union negotiator refusing to sign until Hollywood studios make concessions on AI protections.
Chinese court to rule Replacing a worker with AI does not legally justify firing him, ordering the tech company to pay wrongful severance.
In each newsletter we show how the reader is using AI to work smarter, save time or make life easier.
Today’s workflow comes from the reader Ole T. In Finland:
“I am a new real estate investor, and what I struggle with is market insights, as in our province, there is no transparency or proper historical information available. With Gemini Pro, I have created a concept that analyzes market information.
The tool displays all relevant renewal history and future needs and can also create an account for investment. I then refined the concept using several free AIs (Claude, Codex), one prompt per day, to get a product that suited my needs.”
How do you use artificial intelligence? Tell us here.
That’s all for today!
Before you go, we’d love to know what you thought of today’s newsletter to help us improve The Rundown experience for you.
Rowan, Joey, Zack, Shubham and Jennifer – the humans behind The Rundown
