Asymmetry of Verification and Verifier's Law (jasonwei.net)

1 points by ajuhasz 33s ago 0 comments

Ask HN: Python developers at big companies what is your setup?

2 points by ravshan 1m ago 0 comments

Why LLMs Struggle with Analytics (tinybird.co)

1 points by _peregrine_ 1m ago 0 comments

How to Manage Multiple Claude Code Accounts? (gist.github.com)

1 points by KMJ-007 2m ago 0 comments

1KB JavaScript Numbers Station (shkspr.mobi)

2 points by blenderob 3m ago 0 comments

Are Diamonds Even a Luxury Anymore? De Beers Reckons with Price Plunge (wsj.com)

3 points by randycupertino 3m ago 1 comments

Reverse Proxy Deep Dive: Why HTTP Parsing at the Edge Is Harder Than It Looks (startwithawhy.com)

1 points by miggy 3m ago 0 comments

From REST to Reasoning: A Journey Through AI-First Architecture (oreilly.com)

1 points by almost-exactly 3m ago 0 comments

We built an entire newsroom with LLMs – then we showed the print to the real one (medium.com)

2 points by madkerpe 4m ago 0 comments

Auto-convert multimodal data into ML-ready datasets (github.com)

1 points by leandrenash 5m ago 2 comments

Go Allocation Probe (scattered-thoughts.net)

2 points by blenderob 5m ago 0 comments

Bacon – Bicameral Agent with Compute-Aware Orchestration and Navigation (github.com)

1 points by ycombiredd 6m ago 1 comments

A big problem for Tesla isn't getting much attention (cnn.com)

2 points by breve 6m ago 0 comments

Russia using children to design and test its military drones investigation finds (theguardian.com)

2 points by c420 7m ago 0 comments

I displayed an open graph image and had to pay how much? (alistairshepherd.uk)

2 points by OuterVale 7m ago 0 comments

Explaining Stagnation in the College Wage Premium (frbsf.org)

2 points by littlexsparkee 8m ago 0 comments

The Sounds of MRI (2024) (larsonlab.github.io)

2 points by dcminter 9m ago 1 comments

New federal data show H-1B and H-4 visas now capture 82% of all new US tech jobs (twitter.com)

3 points by DonnyV 9m ago 1 comments

What to expect from Debian/trixie (michael-prokop.at)

1 points by edward 9m ago 0 comments

Enshittification and the Bitterness of Billionaire Bros (paulkrugman.substack.com)

1 points by PaulHoule 9m ago 0 comments

Landmark study flips decades of cholesterol panic aimed at eggs (newatlas.com)

2 points by Brajeshwar 9m ago 0 comments

Dystopia or utopia? The futuristic visions driven by billionaires (abc.net.au)

1 points by Brajeshwar 9m ago 0 comments

Conversations between LLMs could automate the creation of exploits, study shows (techxplore.com)

1 points by Brajeshwar 9m ago 0 comments

Killing the Mauna Loa observatory over irrefutable evidence of increasing CO2 (theregister.com)

2 points by rntn 10m ago 0 comments

The world is winning the war on cancer (economist.com)

3 points by bookofjoe 10m ago 1 comments

Social network and sociodemographics associated with steroid use intention (sciencedirect.com)

1 points by PaulHoule 11m ago 0 comments

AI Could Have Written This: Birth of a Classist Slur in Knowledge Work (dl.acm.org)

1 points by gpi 12m ago 0 comments

The Limits of Agency (Human or Otherwise): Free Will Is Overrated (medium.com)

1 points by hazalmestci 12m ago 0 comments

Show HN: OneArch – AI-assisted architecture docs with ADRs and diagrams (theone.archi)

1 points by nick-laptev 15m ago 0 comments

A non-AI project (for engineers who lift) (apps.apple.com)

1 points by dnignzlz 15m ago 1 comments

Justice Department official plans meeting with Ghislaine Maxwell (theguardian.com)

2 points by belter 16m ago 0 comments

Advising is the next step for LLMs, a change in how they answer questions

1 points by Taikhoom10 17m ago 0 comments

Show HN: Browser AI – Use Playwright Without Blowing Up Your Context Window

2 points by chw9e 18m ago 2 comments

Ask HR: LLMs can't solve basic logic problems, much less code

1 points by xchip 19m ago 0 comments

A Friendly Introduction to SVG (joshwcomeau.com)

2 points by thunderbong 20m ago 0 comments

Import SQL data on an existing server (docs.uxwizz.com)

2 points by XCSme 21m ago 1 comments

Vibe coding a Game Boy emulator (sanjeetnd.com)

2 points by blueblazin 22m ago 0 comments

AI summaries reduce user link clicks on Google search: Pew (pewresearch.org)

1 points by jmsflknr 23m ago 0 comments

Active but still gaining weight? Duke scientists reveal why (sciencedaily.com)

2 points by FollowingTheDao 23m ago 2 comments

Show HN: SandCrab – An AWS S3 GUI for macOS (sandcrab.io)

1 points by __zayne__ 24m ago 0 comments

Ask HN: Will the IMO start grading style to combat overly long AI solutions?

1 points by amichail 26m ago 0 comments

AI Observability Tool/MCP Servers Has No Real Model of Your System (signoz.io)

4 points by elza_1111 27m ago 1 comments

Ask HN: What are your thoughts on the impact of AI on programming careers?

4 points by HisoMorow 28m ago 2 comments

Firefox 141.0 Release Notes (firefox.com)

2 points by nateb2022 28m ago 0 comments

Disrupting active exploitation of on-premises SharePoint vulnerabilities (microsoft.com)

1 points by gnabgib 29m ago 0 comments

One Man's Lifetime Reading Log (what-dan-read.com)

1 points by rdmullins1978 30m ago 0 comments

It's "frighteningly likely" many US courts will overlook AI errors, expert says (arstechnica.com)

2 points by pseudolus 31m ago 0 comments

Bastard Chairs (photomichaelwolf.com)

2 points by NaOH 31m ago 0 comments

Microsoft Says Chinese Hackers Are Exploiting SharePoint Flaws (bloomberg.com)

4 points by wslh 33m ago 4 comments

Postgres schema changes without downtime? pgroll hits 5k stars (xata.io)

3 points by gulcin 33m ago 0 comments

g-AMIE: Towards a Safer Medical AI

1 oldfuture 1 7/22/2025, 7:19:20 AM arxiv.org ↗

Comments (1)

oldfuture · 7h ago

Guardrailed AMIE (g-AMIE) from DeepMind is an AI system that does:

-patient interviews to gather medical history

-never shares medical advice or diagnosis if not reviewed and approved by licensed doctors

apparently g-AMIE followed safety rules 90% of the time compared to 72% for human doctors