Topic · 35 posts

Inteligência Artificial

How AI chatbots manipulate you and undermine your privacy

At the risk of being among the first victims of artificial intelligence in a potential machine uprising, I keep my interactions with today’s generative AIs (or AI chatbots) strictly transactional.

I open the site, ask or request what I need, get the answer, close the site. No names, no pleases, no thank-yous, no small talk. I avoid anthropomorphizing them as much as possible. I treat them for what they are: statistical machines churning out words that make sense, not a new form of sentient life — at least, not yet.

Keeping a transactional relationship with AI chatbots is, for me, a way of keeping the line between us clearly drawn, to ward off an unlikely — but not impossible — “AI psychosis,” the kind of spiral where someone genuinely believes the AI is alive.

A newly published report by the Center for Democracy and Technology (CDT), authored by researchers Ruchika Joshi, Adinawa Adjagbodjou and Michal Luria, gave me further reasons to stand by that approach.

(more…)

At the opening of Build 2026, Microsoft lived up to the nature of the event — focused on…

At the opening of Build 2026, Microsoft lived up to the nature of the event — focused on developers — and announced, among other things, a slew of updates to make Windows 11 more appealing to this audience. The linuxification of Windows is moving forward at full steam, now with native coreutils and container support in Windows Subsystem for Linux (WSL). Beyond that, plenty of things involving AI “agents” — the big theme this year — including native support for OpenClaw.

How to disable WordPress 7.0’s new AI features

Released this Wednesday (20th), WordPress 7.0’s headline feature was supposed to be real-time collaboration. But in the final stretch, after beta versions, Matt Mullenweg, the project’s leader, postponed the feature and promoted LLM integration to be the version’s centerpiece instead.

(Matt’s leadership was a shitshow, with delays, new demands with unrealistic deadlines, and as always, plenty of finger-pointing. “What’s wrong with WordPress?”, asks the only person with the power to change WordPress’s direction.)

If you don’t see the value in or are wary of an infusion of generative AI on your website, blog, or online store (a plausible concern, I must say), fortunately there’s a way to disable all this new AI stuff. If you have access to the wp-config.php file, add this line:

define( 'WP_AI_SUPPORT', false );

If the file is beyond your reach, install the Turn Off AI Features plugin.

You can’t imagine my relief watching this drama from the outside. In early May, we migrated this Manual do Usuário to ClassicPress, a lifeline that I discovered back in 2022 and which, six years later, remains active and better than ever. How can you not love the idea of free software?

Google presented “AI agents” capable of doing people’s work in…

Still thinking about Google I/O…

Google presented “AI agents” capable of doing people’s work in countless scenarios, none of them very believable for real people. (Who needs AI to organize a neighborhood block party?) At TechCrunch, Sarah Perez writes a solid critique of this questionable utopia that Google is trying to sell us.

On my end, I think we’re living a “Groundhog Day” moment of that Google Now feature from ~2012, which would notify you via push notification that your flight’s gate had changed. As if that information wasn’t already on your face the whole time at the airport. For 15 years Google has been using different technologies, each increasingly complex and expensive, to try to solve problems that no human being has ever actually had.

Google wants to be the interface for the web

The opening of Google I/O this year (35-minute video recap) showed a Google less shy about transforming the web into raw material for its AIs.

Search will become (even more) a souped-up ChatGPT, and also the checkout counter for every online store, and YouTube will use video clips to create answer pages.

Notice that in all these announcements, Google/YouTube transforms itself into a curator and interface for web content, without attribution or with minimal credit given. You won’t visit websites anymore; you’ll visit Google — and you’ll stay there. On the flip side, whoever calls themselves a “content creator” is actually a content supplier for platforms. It’s always been that way for Meta (Instagram), TikTok, and YouTube; now it’s also the case for Google, even if unintentionally (myself included!).

All of this amounts to a complete betrayal of everything Google once stood for (and was) back then. Maybe people like this new incarnation because the web, poor thing, has long suffered from the pernicious incentives and destructive influence of Google itself, but nothing is so bad that it can’t get worse.

By the way: Manual do Usuário project offers an instance of SearxNG, a metasearch engine that displays ten (or more) blue links to actual websites. It’s free. Use it and spread the word.

Nvidia announces an AI filter for games, and gamers go wild

Something funny happened in the world of video games.

Nvidia realized that its graphics chips are also good for gaming, not just AI, and announced a technology called DLSS 5 that uses their power to, in the company’s words, “infuse pixels with photoreal lighting and materials to bridge the gap between rendering and reality.” With generative AI, obviously.

The public reaction was quite negative, even on Nvidia’s official YouTube video. DLSS 5 kind of turns the look of games into AI-generated videos. More photorealistic than the original, yes, but the criticism is that the technology interferes with the art of games.

The DLSS 5-on graphics do indeed look like AI-generated videos, or those artificial filters on Instagram and TikTok, but it’s not as if the originals, without the effect, were much better. That intrigued me a bit more than DLSS 5 itself. I haven’t played video games in ages, and these PlayStation 5 and Xbox Series X graphics don’t seem all that different from what I remember of video games from two or three generations ago. Given that, I think I prefer Nvidia’s AI-filtered version…?

This article from Gizmodo has some comparisons and videos.

Jensen Huang, CEO of Nvidia, was unfazed by the widespread criticism. During a Q&A session at Nvidia GTC, the company’s AI-focused event, when asked by a reporter, Jensen began his response saying that “Well, first of all, they [critics] are completely wrong” (video).

AI kill switch comes to Firefox

Firefox 148, released on Tuesday (24th), features a kill switch for the browser's AI features. Below it are selective controls for specific features. All of them are located in a new area in the settings (about:preferences), called “Artificial Intelligence Controls.”

The AI industry doesn’t take “no” for an answer

For days now, I've had a quote from David Bushell stuck in my head:

Has anyone else noticed that the AI industry can’t take “no” for an answer? AI is being force-fed into every corner of tech. It’s unfathomable to them that some of us aren’t interested.

David complained about receiving communications from Proton offering Lumo, its generative AI, even though he had expressly indicated that he did not want to receive such messages. The worst part is that Proton, instead of owning up to its mistake and apologizing, insisted on absurd justifications to say that there was no mistake. It only gave in when an executive got involved, and only after the post went viral.

(more…)

For this [AI] not to be a bubble, by definition it requires that the benefits of this are much more evenly spread. I think a telltale sign of if it’s a bubble would be if all we’re talking about are the tech firms. If all we talk about is what’s happening to the technology side that then that’s just purely supply side.

Bald man wearing glasses, smiling.Satya Nadella,
CEO of Microsoft

I have some bad news for you, Satya…

In another excerpt from the interview, picked up by Pivot to AI, Nadella says that companies need to reorganize around AI to learn, in practice, how to use it in business. Destroy what is working to learn an innovation that might help them… do what they were doing before?

Dell and Microsoft, marketing geniuses

Remember when Warner Bros. changed the name of its streaming service from HBO Max to Max, and then less than a year later back to HBO Max? Or when the richest person in the world proved that money has no relation to intelligence and threw the “Twitter” brand in the trash? Marketing geniuses!

Maybe it's late capitalism, maybe it's a side effect of new drugs circulating among the bigwigs of the world's most powerful companies. Or maybe it's just plain stupidity. The fact is that the practice is spreading, and fast.

In early 2025, Dell revamped its line of notebooks and retired traditional names, including XPS, perhaps the most recognizable after Apple's MacBook. The goal was to simplify. No one understood anything.

Cut to 2026, and Dell announced at CES, to no one's surprise, that it will return to using the XPS brand. They look nice.

Running behind is Microsoft. When accessing the office.com website, we come across this gem (my highlight):

Welcome to the Microsoft 365 Copilot app

The Microsoft 365 Copilot app (formerly Office) lets you create, share, and collaborate all in one place with your favorite apps now including Copilot.*

Just imagine, replacing the brand that has been synonymous with productivity apps for three decades with… Copilot, a slop generator that people generally dislike and only use because their employer requires them to.

Good for us. The less we associate critical software with big tech brands, the better. Long live Microsoft 365 Copilot — or whatever other weird name Word, Excel, and company end up with in the future.

2025 has been a disaster for Windows 11 
windowscentral.com

Zac Bowden wrote a long article stating that Windows fans (the author's definition) have been sold a “disastrous 2025 for Windows 11.” I haven't used Windows in many years and have barely touched version 11, so I read it with extra attention.

Anyway, I'm sure you can guess the most obvious problem with Windows 11 in 2025:

(more…)

The fact that people are unimpressed that we can have a fluent conversation with a super smart AI that can generate any image/video is mindblowing to me.

A picture of Mustafa Suleyman.Mustafa Suleyman
CEO da Microsoft AI

Mustafa Suleyman’s outburst is a reaction to Windows 11 users’ criticism of another Microsoft exec, Pavan Davuluri, who said Windows is evolving into an “agentic” operating system.

“Agentic” is a euphemism tech execs use for “AI‑stuffed software that doesn’t work properly.” In the case of Windows 11, for example, Microsoft warns that AI “agents” can, among other things, install malware and expose private data. Yay…?