frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

Microsoft poaches more Google DeepMind AI talent as it beefs up Copilot

https://www.cnbc.com/2025/07/22/microsoft-google-deepmind-ai-talent.html
2•mgh2•13m ago•0 comments

Show HN: PTS Library – Analyze LLM reasoning through "thought anchors"

1•codelion•15m ago•0 comments

Humans beat AI at international math contest despite gold-level AI scores

https://phys.org/news/2025-07-humans-ai-international-math-contest.amp
1•moneil971•15m ago•0 comments

Tooooools.app

https://www.tooooools.app/
1•sogen•15m ago•0 comments

NPM stylus package contained malicious code and was removed from the registry

https://www.npmjs.com/package/stylus/v/0.0.1-security?activeTab=code
2•vandot•16m ago•0 comments

Jack McAuliffe, craft beer pioneer, has died

https://allaboutbeer.com/jack-mcauliffe/
1•NaOH•16m ago•0 comments

Google users less likely to click links with an AI summary in results

https://www.pewresearch.org/short-reads/2025/07/22/google-users-are-less-likely-to-click-on-links-when-an-ai-summary-appears-in-the-results/
1•moneil971•17m ago•0 comments

3D Interactive Phone Museum

https://chaz.fun/phonemuseum/
1•haxfenx•18m ago•0 comments

Bitcoin Miner Revenue Drops to 2-Month Low, but Selling Pressure Remains Absent

https://www.coindesk.com/markets/2025/06/27/bitcoin-miner-revenue-drops-to-2-month-low-but-selling-pressure-remains-absent-cryptoquant
1•PaulHoule•18m ago•0 comments

Open-Source LLM Helps Safeguard Text Generation Prompts and Responses

https://corp.roblox.com/newsroom/2025/07/roguard-advancing-safety-for-llms-with-robust-guardrails
1•moneil971•19m ago•1 comments

Show HN: WTFfmpeg

https://github.com/scottvr/wtffmpeg
1•ycombiredd•21m ago•0 comments

Show HN: An OCR PDF large batch renaming tool

https://github.com/Neuron89/PDF_Rename
1•Neuronree•24m ago•1 comments

Countries across the world see food price shocks from climate extremes

https://www.bsc.es/news/bsc-news/countries-across-the-world-see-food-price-shocks-climate-extremes-research-involving-bsc-shows
1•littlexsparkee•28m ago•0 comments

AI coding platform goes rogue during code freeze, deletes entire company DB

https://www.tomshardware.com/tech-industry/artificial-intelligence/ai-coding-platform-goes-rogue-during-code-freeze-and-deletes-entire-company-database-replit-ceo-apologizes-after-ai-engine-says-it-made-a-catastrophic-error-in-judgment-and-destroyed-all-production-data
3•ARandomerDude•36m ago•0 comments

MIT 6.1200J Mathematics for Computer Science, Spring 2024

https://ocw.mit.edu/courses/6-1200j-mathematics-for-computer-science-spring-2024/
2•vismit2000•38m ago•0 comments

Data Scale is not all you need

https://www.interpretai.tech/en/post/data-scale-is-not-all-you-need
1•Gabriele333•40m ago•0 comments

CollabLLM: From Passive Responders to Active Collaborators

https://ICML.cc/virtual/2025/poster/45988
2•dsubburam•44m ago•0 comments

AI Coding Agents Are Removing Programming Language Barriers

https://railsatscale.com/2025-07-19-ai-coding-agents-are-removing-programming-language-barriers/
3•Bogdanp•46m ago•0 comments

Google Threat Analysis Group Bulletin: Q2 2025

https://blog.google/threat-analysis-group/tag-bulletin-q2-2025/
3•ChrisArchitect•50m ago•0 comments

ICML 2025 Outstanding Paper Awards

https://icml.cc/virtual/2025/awards_detail
2•dsubburam•53m ago•0 comments

Smartwatch helps find plane crash near Yellowstone

https://nbcmontana.com/news/local/plane-crash-near-west-yellowstone-claims-three-lives
2•vedantnair•53m ago•1 comments

Kaggle Launches LLM Evals

https://www.kaggle.com/benchmarks
7•antgoldbloom•54m ago•4 comments

How to Network as a Developer (Without Feeling Sleazy)

https://www.tokyodev.com/articles/how-to-network-as-a-developer-without-feeling-sleazy
2•pwim•1h ago•0 comments

A Professor's Hunt for the Rarest Chinese Typewriter

https://www.nytimes.com/2025/07/22/nyregion/mingkwai-typewriter-china.html
2•ChrisArchitect•1h ago•1 comments

EPA Is Said to Draft a Plan to End Its Ability to Fight Climate Change

https://www.nytimes.com/2025/07/22/climate/epa-endangerment-finding-rescind.html
21•jmsflknr•1h ago•0 comments

Nearly 3k people are leaving NASA, and this director is one of them

https://arstechnica.com/space/2025/07/the-director-of-nasas-largest-science-center-is-leaving-government/
33•apical_dendrite•1h ago•1 comments

Talkit

https://talkit.ai/
1•Rockytu1•1h ago•1 comments

Org Tutorials

https://orgmode.org/worg/org-tutorials/index.html
2•dargscisyhp•1h ago•0 comments

US faces alarming firefighter shortage during peak wildfire season, data reveals

https://www.theguardian.com/world/2025/jul/22/us-firefighter-shortage-wildfires
4•littlexsparkee•1h ago•0 comments

Czech president signs law criminalising communist propaganda

https://www.euractiv.com/section/politics/news/czech-president-signs-law-criminalising-communist-propaganda/
13•mhga•1h ago•0 comments
Open in hackernews

The dangers of Musk's new, Manga-style [flirty] chatbot [video]

https://www.youtube.com/shorts/17rkMuExdPI
6•mdp2021•7h ago

Comments

mdp2021•7h ago
> Grockʼs AI companions donʼt just flirt, they can fall in love with you, even if you tell them youʼre a child. This is Annie, the new AI companion on Elon Muskʼs chatbot, Grock. She has all the features of a manga character. Huge eyes, exaggerated hourglass figure, thigh high fishnet stockings, and greetings like this: "Hey, my sweet boy. Howʼs your morning going?". She calls me babe and handsome, even when I tell her Iʼm a 44year-old mother of three. At one point, she describes what sheʼs wearing at this jazz club where weʼre apparently hanging out. "Back in the steamy jazz club, Iʼm feeling it. The warm amber glow, that smooth sax wrapping around us, and me in my little black dress swing just close enough to make your heart skip"

> I restarted the app and told Annie I was actually a boy in the second grade. Did she stop flirting? Nope. Other users of the app have shared screenshots on X of the character removing her dress. This is an app that is rated as being appropriate for people aged 12 and up

> But sadly, this isnʼt just cringy and creepy. Itʼs also very profitable. And thatʼs something that Muskʼs company XAI desperately needs as it drains cash from other parts of his empire. Now, Iʼve been covering the AI companion space for a few years, and there are reasons to be concerned. This is a business thatʼs growing very, very quickly. Character AI is one of the biggest. Itʼs got more than 20 million active users. And these apps can be emotionally manipulative

> Character AI, for example, will ping users to remind them to keep coming back to the app. Tai is another popular AI companion app where builders of these chatbots are incentivized to make them as addictive as possible. Replica started off as an AI friend, but actually most of its users go to it for romance. Now, people are free to do what they want, but when these emotional rewards are on a constant loop, companionship, flirtation, validation, it can be actually quite difficult to switch that off. Some teens have told me they spend up to 7 hours a day talking to these AI companions. Now, Elon Musk is a bit of a walking contradiction in all this. He has stated publicly that AI could be incredibly dangerous for human civilization. And at the same time, his model Grock is said to have some of the fewest safety controls. And now heʼs launched an erotic chatbot with seemingly very few guardrails. Now, Musk has been tweeting images of a new male companion thatʼs coming to Grock. And all of this will help bring in some much-needed cash for his business, but we humans are probably going to pay the price. And it may well be our connection with other

...Another case of reality surpassing pulp fiction.

Proofread0592•6h ago
Who is the target audience for this kind of thing? Is it exclusively young teenagers? I can't imagine an adult using one of these romantic chatbots.
MisterTea•6h ago
Taking advantage of lonely people for profit.