Mozilla said Firefox will have a “kill switch” that disables all AI features in the browser. Someone witty quipped it will be the “Uninstall” button.
2025
The iPhone 16e is good, actually
MKBHD gave the “Burst of the Year” award to the iPhone 16e, Apple’s entry-level phone released in the first half of 2025 to replace the tired (but beloved) iPhone SE. The Basic Apple Guy, in his annual Apple product releases tier list, placed the iPhone 16e in tier C.
Those two examples show the “cheap” iPhone didn’t have the impact people expected because… well, it’s not that cheap. (And it has no MagSafe. Fix this on the 17e, Apple?)
The iPhone 16e’s pricing is odd. It costs USD 599, which is only about 25% cheaper than the much better iPhone 17 at USD 799. (Or −27.7% if you consider the unlocked price of USD 829.)
Note that most of the criticism of the iPhone 16e comes from US-based outlets and commentators. The 16e doesn’t really feel like a product designed for that market.
Home Assistant Voice Preview Edition — the open alternative to Alexa and Siri for controlling smart homes
When it comes to smart speakers, Amazon has Alexa, Apple has the HomePod, and Google has Nest. If you want something private — that runs locally — to control your home, there weren’t many alternatives.
Or there weren’t until now. To fill that gap, Nabu Casa, the sponsor of the Home Assistant open source project, released the Home Assistant Voice Preview Edition.
I bought six of these to replace six HomePods I had scattered around the house. After using them for a while, the question is: can you trust this for everyday use, or is it better to wait for a release without “preview” in the name?
It’s possible that Time magazine’s Person of the Year selection — which in this issue named “the AI architects” (i.e., big-tech AI CEOs) — was chosen by an AI. First clue: calling several people the otherwise singular “Person” of the Year. Second and stronger clue: only something as dumb as an AI would pick that bunch of clowns as Person(s) of the Year.
How to get found by recruiters on LinkedIn
Editor’s note: Job openings are a recurring topic in the Manual supporters’ WhatsApp group. In a recent conversation there were so many useful LinkedIn tips that we ended up with a kind of playbook for doing well on the platform. Thanks to everyone who contributed, especially Marcia who steered the discussion, Caique who saved the thread, and Paulo, who condensed many messages into this concise text.
Nobody likes LinkedIn, but a lot of people need to be there to apply for roles at companies that, if you’re lucky, will send an automatic message telling you they decided to move forward with another candidate. Even so, it’s worth trying to understand how this corporate theme park works and how it can (yes, it’s possible) help you land a less miserable job.
Is “green AI” even possible?
Ecosia, the search engine that directs its profits to climate action, asks whether there’s a “green” alternative for generative AI in the title of this post — and in the same post’s headline answers that “the world’s greenest AI is here.”
The “greenest AI,” that paradox, is the one Ecosia just launched integrated into its search. The implementation mirrors Google’s: a summary (“AI Overviews”) above results and a one‑click away chatbot. The obvious difference is a button to disable AI. The less obvious — and crucial — claim is that:
As a not-for-profit company, we can afford to do things differently. AI Search uses smaller, more efficient models, and we avoid energy-heavy features like video generation altogether.
Details are missing (which “smaller, more efficient models”?) and the example is odd (no search engine with AI currently offers video generation).
I’m not against generative AI — that would be hypocritical, since I use it occasionally. (And on another search engine, DuckDuckGo’s.)
That’s why headlines like this and Mozilla pouring everything into AI for Firefox leave me uneasy. No matter the appeal, it seems odd for a service whose raison d’être is climate action to jump on a wave so energy‑hungry it’s reviving nuclear plants and prompting big techs to drop carbon‑reduction promises.
The “greener AI,” ultimately, is not having AI at all. Ignoring it — at least until adoption stabilizes and environmental impacts are better understood — would be coherent and even a differentiator in a sea of companies adding AI for the sake of it.
I tried to build a WhatsApp bot. Meta banned me before it left the drawing board
Part of most people’s learning curve when they start coding is trying projects that, with luck, might become useful to themselves and others. In May I wanted to get a better handle on the WhatsApp API. I set up my local environment, logged into Meta’s developer platform and started poking around.
I grabbed a test number the platform provided, sent a “Hello World” and tried a few basic commands for the architecture I was sketching. A few weeks later I had to set the project aside.
When I tried to pick it up again a few weeks ago, I was alarmed by the message on Meta’s developer page:
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.
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…?
Reply to Anil Dash, re: I know you don’t want them to want AI, but
I was flattered to get a reply from Anil Dash to my post about the backlash Mozilla faces for its plan to add AI to Firefox. I’ve read Anil for a long time and admire his work.
That said, I respectfully disagree with his arguments — here’s why.
I think nobody wants AI in Firefox, Mozilla
Mozilla is developing a built‑in AI assistant for Firefox that will be offered as a third browsing mode alongside Normal and Private tabs. They’re calling it “Window AI.”
Details are still scarce. Based on Mozilla’s official announcement on Thursday (13th), it looks like a deeper implementation than the existing sidebar that gives access to third‑party chatbots (ChatGPT, Gemini, Copilot, etc.). The post stresses the feature will be opt-in and that the user “is in control.”
There’s a waitlist to try the feature and a Mozilla forum thread inviting people to “help shape” the initiative.
How should open source software projects handle AI‑generated code?
The excellent KeePassXC, an offline, open‑source password manager, is at the center of a controversy over the use of AI‑generated code after the project’s collaboration policy and README added this paragraph:
Generative AI is fast becoming a first-party feature in most development environments, including GitHub itself. If the majority of a code submission is made using Generative AI (e.g., agent-based or vibe coding) then we will document that in the pull request. All code submissions go through a rigorous review process regardless of the development workflow or submitter.
Users and critics backlash was so intense that on Sunday (9the) one of the project maintainers, Janek Bevendorff, published a post on the official blog detailing their stance on AI‑generated code.
Facebook and Instagram are paradises for scammers, reveal Meta’s internal documents
Reuters shed light on Meta’s lucrative business built on selling fraudulent ads on its platforms — Facebook and Instagram. Internal company documents obtained by the news agency show that 10.1% of Meta’s 2024 revenue, or US$ 16 billion, came from fraudulent/scam ads.
A December 2024 document shows Meta running an average of 15 billion fraudulent ads per day. Those add to 22 billion pieces of suspicious “organic” content (unpaid) — from hacked profiles offering crypto schemes to promises of miracle cures in groups, and fake listings on Facebook Marketplace.
A few months ago, youtubers reported unsolicited interventions by Google to “improve” their videos with generative AI. It appeared to be a test; now it’s official.
Channel owners can disable this feature in Studio: go to Settings, Channel, Advanced settings and uncheck the two options under Video quality enhancements. For viewers, Google’s suggested workaround is to change the resolution in the player’s settings.
2005: Don’t trust anything you see on Wikipedia.
2025: Don’t trust anything you see online unless you saw it on Wikipedia.
Jason Lefkowitz
@jalefkowit@vmst.io
About the password leak of 183 million Gmail accounts
In the same vein as the “phones that will stop running WhatsApp” beat, Brazilian news sites seem to have found a new evergreen click source for tech desks imported from Forbes’: millions of leaked Gmail passwords.
There is, in fact, a database of that type circulating online, created by an undergraduate student in the United States. Troy Hunt, who runs Have I Been Pwned, a breach repository, analyzed the data and found that “only” 8% of the passwords — about 14 million — are new. That makes sense, given the database was glued together by aggregating entries from multiple sources and prior breaches.
The main takeaway from a story like this isn’t “your Gmail password may have leaked,” but rather that “any of your passwords could leak at any time.” Not to spread alarm, but to encourage awareness of good digital security practices.
Which ones? For this situation, mostly these two:
- Use a password manager. It makes easy creating and retrieving strong, unique passwords for each service.
- Enable two‑factor authentication (or two‑step verification). It can be integrated with the very same password manager for easier adoption. In a breach, the second factor blocks unauthorized access even if someone has your password.
You can check whether your passwords have leaked by entering your email at Have I Been Pwned. If it shows up, there’s no need to panic: change the password and enable a second authentication factor. Google explains how to do this for Gmail.