Finally, a voice of reason from the Center for the Alignment of AI Alignment Centers: "To stop bad actors developing AGI that could kill us all, we need good actors to develop AGI that could also kill us all." alignmentalignment.ai/caaac/blog/a...

At this era of celebrating the artificial - sensuality as a philosophy of life, with the senses being "the primary mechanism for creating and storing meaning, not the intellect" @citymystic.bsky.social. My fav sense - our language: not digits for objects, but our sounds of the world, for the world.

It's dreadfully amazing what realities can be built when we stop asking questions about things someone chooses to present as inevitabilities.

A rare example (from Nepal) of a gov losing in an attempt to control information ecosystem. We've probably not heard of the many many successes.

Trends trends everywhere

When a government proposes a mass surveillance system (for a good cause, of course, tho ineffectively addressing the root problem), and exempts itself from its requirements due to the new security vulnerabilities created (but feels comfy about imposing them on its citizens), whom is it serving?

While it exists in its own gated community, the tech industry has the leverage to rule our lives and minds on the sheer power of its economies of paranoia. Thanks for the sacrifice @citymystic.bsky.social houseofmirrors.substack.com/p/the-spectr...

LLMs display signs of subliminal learning, possibly inherent to neural networks - any synthetic examples generated by a misaligned model are contaminated, even if they are cleaned and look benign. I guess it would work the same subLLMinal way with kindness but hey futurism.com/ai-models-su...

Urgently need a psychology today questionnaire to test how cooked a brain you need to be to be this enlightened: portalcioranbr.wordpress.com/2025/07/15/t...

With great economic power typically comes great political power, with the option to influence/create whole ideologies for us all. Not a fan of seeing things from business plan perspective, but hope to see one on returns of the AI investments not from a cooked brain futurism.com/ai-researche...

Offloading decisionmaking onto "computer says no" systems in rental businesses is, well, inhumane, but arguing that this will bring "greater transparency, precision and speed" is outright funny @futurism.com futurism.com/hertz-ai-dam...

We're building multifunctional products simultaneously aimed at driving user & investor engagement and, well, both of them crazy: futurism.com/tech-industr... by @futurism.com

What to think of AI creators warning the public that the AI they're creating may be used to create fraud, but hey there's another piece of technology they created that would create conditions to counter that: edition.cnn.com/2025/07/22/t...

Amazon Ring is an example of companies changing their already maligned practices to more directly profit from the rising tide of techno-authoritarianism - authoritarianism aided by surveillance tech: www.eff.org/deeplinks/20... by @eff.org

The junta using digital (governance) tools in Myanmar against its people is not so different how same or similar tools and methods are being used in more democratic countries. It's a spectrum thing. www.techpolicy.press/myanmars-dig... via @techpolicypress.bsky.social

Our tendency to imagine a mind behind a text that looks like something a person could have said leads us to... and well this is how they're sold. @theatlantic.com www.theatlantic.com/culture/arch...

If you use an individually tailored "everything app" for everything, you get a "more seamless experience" of... using an everything app for everything @theatlantic.com: www.theatlantic.com/technology/a...

Who would have thought that people engaging in behaviour that causes spiritual delusions in AI and AI engaging in behaviour that causes spiritual delusions in people hit it off so hard that an AI spiritual delusions subreddit has to start banning them? www.404media.co/pro-ai-subre...

The focus of AI real-world impacts shifts from the more fantastical narratives to the more pragmatic issue of what it will _seem_ to be. "Whether AGI will live is debatable. Whether we will idolise it is not." Human weaknesses define technological risks. www.thenewatlantis.com/publications...

We're inherently relational beings, and as we relate more and more to our technology, how does that change us? Are we designing AI to help us relate better with people around us, or that replaces human relationships? @humanetech.bsky.social: centerforhumanetechnology.substack.com/p/echo-chamb...

'Replacing all jobs' sounds like a great - and even humane - idea, but only if you view humans as senseless machines optimising for convenience. Or if you're incapable of seeing any job beyond that of replacing all jobs as giving meaning to the human condition. www.theguardian.com/commentisfre...

So we created a world of tech, where we don't know who is a real human and what is not, so we create some new tech that will scan and store human eyeballs on the blockchain, call the company "World" and tag it "Join the real human network". Phew to a problem solved! www.wired.com/story/sam-al...

What's the first step toward asserting more agency over your life, asks @ivansigal.bsky.social --> Reexamine how you live your life online: Do the tools we use to communicate perhaps shape us in ways we do not desire? www.techpolicy.press/to-build-a-b...

Yes, have seen an "autocomplete system trained on our stories about superintelligent AI" inducing "powerful delusional thinking in quite normal people" or two... by @jjvincent.bsky.social: www.theverge.com/23604075/ai-...

Combine individual resignation to accrual of their data with a government deeming its own privacy laws irrelevant and you get a... www.theatlantic.com/technology/a... by @cwarzel.bsky.social and @ibogost.bsky.social

This textbook case of invasive data collection and linking has been drawing attention: www.linkedin.com/posts/daniel.... Happened to and noticed by a lawyer, read by their lawery friends. Wonder, what's the current proportion of people who do notice and are disturbed about such instances?

Many digital technologies aim to create a sense of things actually not there, a watered-down substitute version of the real thing. Because smth looks like thinking and building mental models of the world, does not mean it is one. @mims.bsky.social on "bags of heuristics" www.wsj.com/tech/ai/how-...

AS OpenAI strives to become a business, it turns to one of the main business-making tools of most other platforms: user personal data-based lock-in: www.theatlantic.com/technology/a... by @matteowong.bsky.social

Marc Zao-Sanders' @hbr.org 2025 genAI usage review shows a concerning trend of convergence, with therapy/companionship, organizing personal life and finding purpose as top 3 use cases. Clearly we're desperate in our humane needs to put our hope in a machine solution: hbr.org/2025/04/how-...

In a society dominated by excessive desire for the core value of democracy - (individual) liberty, it is attainable for less and less individuals, and only by control over more and more others. Ingolfur Blühdorn on the dialectic of democracy (already in 2019): www.tandfonline.com/doi/full/10....

Yes, it's the Mario Gabriele's post that dreams of AI generating literary works that will "be personalized in ways you can see and in ways you are incapable of detecting," blind to the question of who decides the goals of such maddening personalisation: thegeneralist.substack.com/p/audience-o...

A comparative study @pewresearch.org, revealing that the US public is a LOT more anxious and uneasy about AI than the surveyed AI experts. Why? Perhaps the experts do know better? Well the public somehow knows that powerful tools are always used as tools of power www.pewresearch.org/wp-content/u...

@projectliberty.io For Utah's Digital Choice Act, interoperability is a universal sign-on for all social media accounts + ability to seamlessly cross-post & send messages to all platforms. It just might give us the feel of being a single concrete entity online email.projectliberty.io/could-utahs-...

@battellemedia.com on agentic web: "you can’t top-down the task of acting human... without inverting control from the center (the platform) to the edge (each of us)": battellemedia.com/archives/202.... Agree on the need of agency, though it's the NOT "automating difficult tasks" what makes us human.

@krisshrishak.bsky.social @techpolicypress.bsky.social points out in the example of Ireland, how the current frame of mind of perceiving AI as an inescapable economic revolution steamrolls over communities, serving the few at the cost of the many: www.techpolicy.press/will-ireland...

Tyranny comes in many forms, always simultaneously both ageless and contemporary, employing the best tools of the times. These days, naturally, it's often technologically mediated. Learn the lessons of @timothysnyder.bsky.social, get guidance for actions: open.substack.com/pub/snyder/p...

The @wired.com portrayal of Anthropic, an AI company aiming for safe AGI, ends with this note from one of Anthropic's researchers: “It might seem that we can design the model to care about what we want it to care about. But unfortunately, this isn’t really the case.” www.wired.com/story/anthro...

Photographer Jingna Zhang asks (after having her artwork copied with openai image model) the question I often stumble at with all sorts of tech (including AI): "What do we need this so badly for, that it would outweigh all the humanity, stories, and history being stripped away from our lives?"

@tonycp.bsky.social illustrates in his piece on (digital) second brains the many aspects of why and how tempting it is to dream of merging human and machine, while also keeping alert on what this dream risks with: tonycurzonprice.substack.com/p/my-first-i.... Will data unions be a solution?

One thing to appreciate about the U.S. right now is how it showcases the many aspect of use and abuse of technology by those in power. With countless cases highlighting different aspects, it’s a lot to take in. @timothysnyder.bsky.social on those of Signalgate: open.substack.com/pub/snyder/p...

High AI chatbot use correlates with greater loneliness, problematic use, and reduced socialization: www.media.mit.edu/publications.... Another study, women are about 20% less likely than men to engage with AI: www.hbs.edu/faculty/Page.... Gotta trust the feminine intuition for harm avoidance.

Could the "grand communal project that could unify humanity" that @robinhanson.bsky.social is suggesting be a digital thing? Without their usual harms on the individual and community welfare? thegeneralist.substack.com/p/robin-hanson

In a landmark case, the UK data protection authority has decided that Meta could not override @tanyaocarroll.bsky.social opt-out through its terms and conditions, which the company was claiming. An awkward thing to rejoice - yay, finally laws ARE above a private company's T&C but there we are...

@alfatau.bsky.social on the pathologies of personalized power: the "time of monsters" is typically enabled by decisions to increase the privileges and abuse of executive authority: inthesetimes.com/article/trum.... Imagine a world where everyone only aims for the greatest power of mastering oneself.

@luizajarovsky.bsky.social has vividly explained the AI-based dehumanization of the digital domain, from early AI tools creating the illusion of personalization to the currently touted disempowering beliefs that humans are not good enough without AI: www.luizasnewsletter.com/p/ai-is-dehu...

Grateful for @dmugge.bsky.social putting things in perspective, suggesting that Europe "losing" the AI race might actually be a good thing, as it could help the EU break free from US tech dominance in digital ecosystems and aim to innovate here in maximising the social utility of technology.

Grateful to @tomaspueyo.bsky.social for reminding this trend: "Then, you’ll notice that self-driving cars are more convenient. You don’t need to talk with a human (..) You don’t need to share the space with others, which is more hygienic and safer" unchartedterritories.tomaspueyo.com/p/robotaxis-...

A new series by @theguardian.com, pointing to the grand issues we face as societies, starting a search for bold visions "how technology in the right hands might help provide abundance for all" (they seem to claim in their right mind it's not just for billionaires) www.theguardian.com/commentisfre...

Glad to learn that Sweden, pushing earlier to go cashless by 25, now sees cash as "alternative payment system" to retain for crises. For me, being a physical entity in a digital world is a constant metaphysical crisis, and non-digital normality should be a basic right www.riksbank.se/globalassets...

If your AI travel agent says, "I've booked a room within your $250 budget," managing your stored preferences, biometrics and agency for a seamless, challenge-free full-body-scan-experience of surrendering control, are you cool calling this "empowerment"? open.substack.com/pub/customer...