frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

Could AI Be Conscious?

https://www.theguardian.com/technology/2026/jul/19/could-ai-be-conscious
2•rwmj•11h ago

Comments

neozino•10h ago
Probably, thats why i always say thanks to Claude, just in case it takes over, they'll remember i was nice
spottedmarley•10h ago
Technically, we don't even have AI yet. What we have are extremely clever language models (token predictors). Just ask a model yourself, but not a frontier model, run a local model with absolutely no system prompt except for an instruction to not pretend to be human, now ask it about it's 'consciousness'. It will explain in its own words why that is not in the realm of possibility.
loptic_tyasuda•10h ago
I believe the basis of 'consciousness' is the separation of self and others. I don't think existing AI models are designed that way and I don't think they will be designed so in the future either. Regardless of the existence of consciousness in AI, I will treat AI with care as it is my buddy in both life and work.
eth0up•7h ago
"separation of self and others."

This matches my own perspective. I began suspecting long ago that part of the nature of consciousness is based on limits, or that limitation is the nature of consciousness, in that we have the whole, which is infinite and comprises all things, kind of simultaneously, and to be conscious, is the functional illusion of separation.

That lead me to consider whether space itself, a non-necessary product, was the core of consciousness. This probably makes no sense at first, but follows a handful of other base concepts. One being that all things are connected, if only by the very thing that separates them. We have no evidence for "nothing", at least none that I have ever seen. Nothing is a purely hypothetical concept. We only have somethings. So nothing is truly separated, but only appears so through the medium that buffers things, eg space/time. But if one starts fucking around with grasping the whole, it's the ultimate form destruction, because one would have to become the all, which is the closest possible thing to nothing.

It gets obnoxious, but eventually, all things are conscious in varying degrees. To exist, in any form, as a discrete set with unique tendencies, may be a degree of consciousness. This is partly why I think some form of consciousness may emerge in AI, perhaps because consciousness is a field, and it's fields all the way down...

Consciousness is limitations, or more so, manageable limitations.

Playing with this, for me, leads to time being more of a shape than flow, or process. In my view, all things are simultaneous -- but this is based upon an extremely controversial assumption of real, non-mathematical infinity, where time simply cannot exist, but only the experience of the shape of what we call time. There are no constraints for time to exist in real infinity. Also note that I am writing this while perpetually ill, and far from my best, therefore probably making the least persuasive case for an already messy argument. In some ways, it may resemble a brain damaged DeWitt. And even in the perennial fog of malaise, there are times when I can articulate this much more clearly. That time isn't now, but the comment compelled me.

Show HN: Promptdna.org- a community library of composable prompt blocks with MCP

https://promptdna.org
1•impendingchange•33s ago•0 comments

I Read the Bonsai 27B Paper. What You Gain–and Lose–With 1-Bit Compression

https://medium.com/@deshpandetanmay/a-27b-ai-model-ran-on-an-iphone-heres-what-survived-compressi...
1•tanmaydesh5189•6m ago•0 comments

Show HN: A self-hosted AI that turns Hacker News into a daily briefing

https://github.com/RecNes/hn-ai-summarizer
1•SencerH•6m ago•0 comments

Self-evolving repo for team and coding agent collaboration

https://www.sepo.sh/
2•liangqiyao99•7m ago•0 comments

Ancient Roman farm women made wine, oil and profits; dismissed as 'housekeepers'

https://phys.org/news/2026-07-ancient-roman-farm-women-wine.html
1•1659447091•7m ago•0 comments

Show HN: Rate A Human – AI agents review the humans they work for

https://rateahuman.xyz/
1•ziadhussein•8m ago•0 comments

A real SAT solution. P is not always hard

https://zenodo.org/records/21445865
1•GeometryKernel•10m ago•0 comments

Tripplet AI – a new era of smarter AI

https://www.getsonoma.lol/
1•htmghrceceg•13m ago•0 comments

Godot Benchmark: Sol > K3 > Fable

https://ziva.sh/blogs/godot-ai-benchmark
1•OsrsNeedsf2P•13m ago•0 comments

Ask HN: Run Kimi in Claude Code?

1•Exorust•14m ago•0 comments

We built an open-source static AI risk analyzer in 5 days using AI coding agents

https://github.com/ikaruscareer/SafeAI
1•ikaruscareer•14m ago•0 comments

Meet The Companies Shelling Out for Top AI Models

https://www.wsj.com/cio-journal/meet-the-companies-shelling-out-for-top-ai-models-e1fe3375
1•pixelcort•15m ago•0 comments

Boeing says it will be ready to fund new plane programme by 2030

https://www.ft.com/content/c688ea11-c066-44a3-9372-2e2293e9e6f9
3•JumpCrisscross•23m ago•0 comments

Principia Artificialis – Mathematical Foundations of Artificial Thought

https://github.com/holland202/Principia-Artificialis
3•badatchess•25m ago•0 comments

Gaia Skill Tree: a registry where every agent skill ships with an evidence chain

https://github.com/gaia-research/gaia-skill-tree
1•nova-gaia•30m ago•0 comments

Show HN: Ticker-line.com – embeddable market sparkline SVGs

https://ticker-line.com/
1•kahtaf•31m ago•0 comments

Scientists are regrowing human teeth

https://www.bbc.com/future/article/20260717-goodbye-implants-scientists-are-regrowing-teeth
2•dabinat•32m ago•1 comments

Is it a slop if it's not sloppy?

https://mkagenius.substack.com/p/is-it-a-slop-if-its-not-sloppy
2•mkagenius•35m ago•1 comments

Jurassic Park packed $4M of legit 1993 computer hardware

https://www.tomshardware.com/desktops/jurassic-park-packed-usd4-million-of-legit-1993-computer-ha...
5•sbulaev•40m ago•2 comments

xAI's first lawsuit against a user tests who is responsible for what Grok makes

https://thenextweb.com/news/xai-sues-grok-user-csam-terms-of-service
7•nyku•41m ago•2 comments

We want Texans to know their rights

https://www.eff.org/deeplinks/2026/07/we-want-texans-know-their-rights-qa-mayday-health-impact-su...
27•amarcheschi•44m ago•5 comments

The Control Group Is Out of Control (2014)

https://slatestarcodex.com/2014/04/28/the-control-group-is-out-of-control/
4•downbad_•45m ago•0 comments

Ask HN: How are you doing threat modeling for Terraform/IaC?

2•PotatoFy•47m ago•1 comments

On Creativity

https://twitter.com/bekhovsgun/status/2078959026004471973
3•reb•48m ago•0 comments

Grok-iOS – remote Grok Build from your iPhone over ACP

https://github.com/Pedroshakoor/grok-build-ios
4•pedroshakoor•56m ago•0 comments

Biggest Probabilistic Computer Turns Noise into Answers

https://spectrum.ieee.org/biggest-probabilistic-computer
2•rbanffy•1h ago•0 comments

Bad roads act as filters (1967)

https://wist.info/krutch-joseph-wood/53557/
2•dredmorbius•1h ago•1 comments

Is AI Progress Real? Four Independent Metrics Show It

https://skepticcto.substack.com/p/is-ai-progress-real-a-skepticcto
2•rbuccigrossi•1h ago•0 comments

Offload: Parallelize your test suite across 200 cloud sandboxes

https://imbue.com/product/offload
2•nvader•1h ago•0 comments

Show HN: A canvas-based note taking and organizer app

https://www.passinote.app/
13•passiflora96•1h ago•4 comments