惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
Microsoft Azure Blog
Microsoft Azure Blog
博客园 - 三生石上(FineUI控件)
WordPress大学
WordPress大学
人人都是产品经理
人人都是产品经理
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
博客园 - 聂微东
Jina AI
Jina AI
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
T
Tailwind CSS Blog
罗磊的独立博客
爱范儿
爱范儿
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
Apple Machine Learning Research
Apple Machine Learning Research
博客园 - Franky
阮一峰的网络日志
阮一峰的网络日志
雷峰网
雷峰网
博客园 - 叶小钗
美团技术团队
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
月光博客
月光博客
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
大猫的无限游戏
大猫的无限游戏
The Cloudflare Blog
Last Week in AI
Last Week in AI
S
SegmentFault 最新的问题
博客园 - 【当耐特】
小众软件
小众软件
Hugging Face - Blog
Hugging Face - Blog
量子位
宝玉的分享
宝玉的分享
V
Visual Studio Blog
博客园_首页
IT之家
IT之家
V
V2EX
腾讯CDC
aimingoo的专栏
aimingoo的专栏
博客园 - 司徒正美
Microsoft Security Blog
Microsoft Security Blog
酷 壳 – CoolShell
酷 壳 – CoolShell
H
Hackread – Cybersecurity News, Data Breaches, AI and More
Blog — PlanetScale
Blog — PlanetScale
I
InfoQ
有赞技术团队
有赞技术团队
J
Java Code Geeks
Recorded Future
Recorded Future
Engineering at Meta
Engineering at Meta
Vercel News
Vercel News
CTFtime.org: upcoming CTF events
CTFtime.org: upcoming CTF events
H
Help Net Security

Ars Technica - All content

Pentagon wants $54B for drones, more than most nations’ military budgets Mozilla: Anthropic's Mythos found 271 security vulnerabilities in Firefox 150 Supreme Court arguments make it clear that FCC fines are "nonbinding" Silo S3 teaser hints at the wasteland's origins Framework's CEO on the RAM crisis and creating a "MacBook Pro for Linux users" Florida probes ChatGPT role in mass shooting. OpenAI says bot "not responsible." Report: Meta will train AI agents by tracking employees' mouse, keyboard use Microsoft removes Call of Duty from Game Pass, lowers subscription pricing Framework Laptop 13 Pro is a major overhaul for the modular, upgradeable laptop Framework Laptop 16 upgrades make it look less like an unfinished prototype Internal emails show how Amazon raises prices across the Internet, lawsuit says Anthropic gets $5B investment from Amazon, will use it to buy Amazon chips CATL's new LFP battery can charge from 10 to 98% in less than 7 minutes AMD Ryzen 9 9950X3D2 Dual Edition review: Tons of cache for tons of dollars What's the deal with spacesuits for the Moon? Will they be ready in time? Loneliness in older adults can often lead to memory impairment Contrary to popular superstition, AES 128 is just fine in a post-quantum world Pentagon pulls the plug on one of the military's most troubled space programs John Ternus will replace Tim Cook as Apple CEO Blue Origin's rocket reuse achievement marred by upper stage failure I’ve fired one of America’s most powerful lasers—here’s what a shot day looks like Great white sharks are overheating US-sanctioned currency exchange says $15 million heist done by "unfriendly states" Man with @ihackedthegovernment Instagram account tells judge, “I made a mistake" Trump picks qualified, normal health leader to head CDC; experts still cautious $25,000 buys plenty of used EVs: Here are some options Satellite and drone images reveal big delays in US data center construction Amazon won’t release Fire Sticks that support sideloading anymore Ridley Scott's post-apocalyptic The Dog Stars drops first trailer Artemis II pilot talks about what it was really like to fly and land in Orion Meta's AI spending spree is helping make its Quest headsets more expensive Rocket Report: Starship V3 test-fired; ESA's tentative step toward crew launch Recent advances push Big Tech closer to the Q-Day danger zone After a saga of broken promises, a European rover finally has a ride to Mars Lucasfilm drops The Mandalorian and Grogu final trailer at CinemaCon Intel refreshes non-Ultra Core CPUs with new silicon for the first time OpenAI starts offering a biology-tuned LLM As they got close to the Moon, Artemis II astronauts were eager to land Mozilla launches Thunderbolt AI client with focus on self-hosted infrastructure Ad firms settle with Trump FTC over claims they boycotted conservative media New Codex features include the ability to use your computer in the background The Ukraine war's deep impact on Metro 2039’s development, story New undersea cable cutter risks Internet’s backbone Microsoft and Stellantis want to use AI to help car owners Gemini can now create personalized AI images by digging around in Google Photos RFK Jr. forces FDA to reconsider 12 unproven peptides after 2023 ban First look: Also's upcoming e-bike disconnects the pedals and wheels Meet the Quantum Kid The race to Shackleton Crater is on—will Jeff Bezos or China get there first? Florida surgeon charged with killing man after removing liver instead of spleen Jury finds Live Nation/Ticketmaster is illegal monopoly that overcharged fans "TotalRecall Reloaded" tool finds a side entrance to Windows 11's Recall database Google releases new apps for Windows and MacOS Boston Dynamics’ robot dog now reads gauges and thermometers with Google's AI Prime Video shows “technical difficulties” sign instead of NBA game in overtime New teaser gives us first look at Godzilla Minus Zero Vulcan woes will "absolutely" be a factor in Pentagon's next rocket competition Adobe takes Creative Cloud into Claude Code-esque territory Good Omens S3 trailer sets up a blessed conclusion Bubble watch: Fashion brand Allbirds pivots hard to become AI services company New 3D map of Universe could solve dark energy mystery What’s the deal with Alzheimer’s disease and amyloid? Blue Origin has a new employee stock plan, but not everyone is happy It's Tax Day, and no one knows how to file for prediction market winnings Ukraine’s military robot surge aims to offset drone risks to humans Sony killing features for antenna, set-top box users of Bravia smart TVs in May Americans ask AI for health care. Hospitals think the answer is more chatbots. NASA chose the right crew to launch a new era of human space exploration Google will begin punishing sites for back button hijacking in June Retro Rewind re-creates the glorious drudgery of working a '90s video store Google shoehorned Rust into Pixel 10 modem to make legacy code safer NZXT agrees to let customers keep their rental PCs in class-action settlement Your tech support company runs scams. Stop—or disguise with more fraud? Sunrise on the Reaping teaser brings us a Second Quarter Quell IBM folds to Trump anti-DEI push, admits no misconduct but pays $17M penalty Slate Auto raises $650 million as production gets closer and closer Meta spins up AI version of Mark Zuckerberg to engage with employees To teach in the time of ChatGPT is to know pain Shock from Iran war has Trump's vision for US energy dominance flailing The Artemis II mission has ended. Where does NASA go from here? AI models are terrible at betting on soccer—especially xAI Grok Four astronauts are back home after a daring ride around the Moon Californians sue over AI tool that records doctor visits New paper argues history, not mantle plume, powers Yellowstone F1 moves a step closer to fixing its 2026 hybrid problem Report: US demands Reddit unmask ICE critic, summons firm to grand jury Microsoft's "commitment to Windows quality" starts with overhaul of beta program "Oobleck" still holds some surprises YouTube increases Premium price again, says 90-second unskippable ads are a bug Oldest octopus fossil found to not be an octopus What leaked "SteamGPT" files could mean for the PC gaming platform's use of AI Here's what to expect from the fiery, 14-minute return of Artemis II Pro-Iran Explosive Media trolls Trump with AI-generated Lego cartoons Dad stuck in support nightmare after teen lied about age on Discord Rocket Report: Chinese version of Falcon 9 fails; Artemis depends on rapid heavy lift Orion helium leak no threat to Artemis II reentry but will require redesign RFK Jr. rewrites CDC panel's charter, opening door to anti-vaccine quacks AI on the couch: Anthropic gives Claude 20 hours of psychiatry Clinical trial shows gene editing works for β-Thalassaemia, too “Negative” views of Broadcom driving thousands of VMware migrations, rival says
The hidden cost of Google's AI defaults and the illusion of choice
Ryan Whitwam · 2026-04-30 · via Ars Technica - All content

AI’ll be watching you

Google says it respects user privacy in AI, but the reality is not so black and white.

Gemini's privacy controls are multifaceted and often confusing. Credit: Aurich Lawson

Gemini's privacy controls are multifaceted and often confusing. Credit: Aurich Lawson

Many people are hoping—nay, praying—that the potential AI bubble will burst soon.

But to hear Google tell it, generative AI is the future, and the company’s products have to change to keep up with the technical reality. As a result, Gemini is seeping into every nook and cranny of the Google ecosystem. Generative AI feeds on data, and Google has a lot of your data in products like Gmail and Drive. What does that mean for your privacy, and what happens if you don’t want Gemini peeking over your shoulder? Well, it’s kind of a mess.

The amount of data Gemini retains depends on how you access the AI, and opting out of data collection can mean running straight into so-called “dark patterns,” UI elements that work against the user’s interest.

This is the future?

Google doesn’t train AI with your data, except when it does

Concerns over how Google uses your private data stretch back long before the generative AI boom. When Google stepped up advertising in Gmail, the company clarified that it doesn’t use the content of your emails to serve ads. Instead, ad personalization (which you can disable) is a global feature that uses your web activity and vital statistics to target you. Most people have come to terms with that, but things get murkier in the realm of AI.

As Google rolls more Gemini features into iconic products like Gmail, it has again had to clarify when it does and doesn’t use your data. In a recent blog post (and associated YouTube Short), Google sought to clarify that your emails are not being piped directly into Gemini. Instead, Gemini gets access to your data for “isolated tasks.” When you interact with Gemini features in Workspace apps like Gmail or Drive, the AI processes your data but does not save it.

“Protecting users’ privacy and control over their data is fundamental to how we develop and deploy AI in Google Workspace,” a Google spokesperson said. “The content you put into Workspace—like your private Drive files—is yours, and when using Gemini in Workspace we do not use that personal content to train our foundational generative AI models.”

So Google doesn’t scan your inbox or documents to train Gemini. Great. But Gemini can use tools to connect to Workspace and other Google products when appropriate, based on your prompt. Google says its AI models can be trained on Gemini inputs and outputs. And guess what those outputs might include—yes, your data.

Gemini outputs can include summaries and snippets of email or files, and that data can then become fodder for AI training. Google says the goal is to train Gemini to be a better assistant, and the way people interact with the bot is a key element, but it tries to “filter and reduce” personal information going into AI training datasets. There’s no way for us to know how well this automated process works, though.

Google is quick to point out that users have control over these features. If you want to keep your private data truly private, you can avoid ever letting Gemini interact with your files or opt out of sharing any data for AI training. Unfortunately, Google doesn’t make it easy to leave Gemini behind.

In a chatbot, darkly

There are a few ways, some of which are more straightforward than others, to ensure that your personal data doesn’t end up in Google’s AI training set. The easiest way is to keep your interactions with Gemini light and impersonal: Don’t allow Gemini to access your other Google apps and stick to temporary chats for anything even remotely sensitive. That way, Gemini won’t have any personal data in outputs to mine. Of course, this makes Gemini much less useful, a familiar theme to those trying to preserve their privacy.

To fully block AI training on your data, you need to turn off a feature called Gemini Apps Activity. This esoteric settings page lets you turn off Gemini history or turn it off and delete your existing Gemini data. If you don’t save activity, Google doesn’t train on it. However, this also means you lose your chat history. So you have to choose between not having your AI chats for future reference or allowing the content of those chats to be used for AI training.

Companies generally won’t admit to designing an interface to manipulate users, but the intent doesn’t determine whether a UI design is a dark pattern. “It doesn’t matter whether it’s intentional or not,” said Marie Potel of Fair Patterns, a startup building AI models that detect dark patterns and predatory design. “What matters is whether the autonomy—the agency—of users is respected and whether the design goes against what users want to do.”

If the only way to opt out of AI training is to permanently disable your chat history, that arrangement doesn’t seem to respect the user’s agency—it’s a forced action. Even finding the right menu to opt out of training can be a chore. There’s a link hiding in the Gemini app settings, but it’s labeled only as “Activity,” and there are also direct links if you search through Google’s support articles.

Interestingly, the Gemini controls are absent from Google’s account privacy settings, where you’d expect to find them. A company representative said there should be a link in the Activity Controls, along with submenus for Android, Maps, Search, Assistant, and more. After checking multiple accounts, we have yet to see a link to the Gemini privacy menu.

Google privacy links UI

Two different Google accounts, no links to Gemini’s privacy menu.

Credit: Ryan Whitwam

Two different Google accounts, no links to Gemini’s privacy menu. Credit: Ryan Whitwam

A toggle that is so vital to user privacy, even one that substantially degrades the experience, should not be so hard to find. Unfortunately, this is not the first time Google has seemingly misplaced a privacy option. Marie Potel noted that hidden and dead links are a dark pattern almost as old as the Internet.

“Google has a history of hiding features, especially privacy settings, in a number of clicks that are absolutely made to deter people from using them,” said Potel.

AI is the new default experience

Accumulating training data is not Google’s only goal with Gemini—it also wants to make generative AI a part of your life. The more people using Gemini features, the easier it is for Google to justify the huge investment in the tech ($185 billion expected in 2026 alone). Naturally, Google’s core products use AI features more aggressively, and it’s assumed that you want them.

In Gmail, Gemini can craft and tweak emails, summarize email chains, organize your inbox, and create AI Overviews of your email. The amount of AI you get depends on whether you’re paying for higher AI limits, but everyone gets some, and these features are constantly expanding. But maybe you don’t want the AI generating hallucinatory summaries in your email. Turning that off is an exercise in frustration because there is no granularity—plenty of features have simple toggles in the Gmail settings but not Gemini.

To disable Gemini features in Gmail, you have to turn them all off through what Google calls “Smart Features,” and there are two ways to do that. One of them nukes a ton of Gmail functionality, and neither is explained very well.

In the Gmail settings, there’s a checkbox toggle for Smart Features that disables Gemini, but it also kills a number of popular features that predate the AI boom. Using this option means giving up inbox filtering (tabs for primary, social, promotions, etc.), Smart Compose, package tracking, and more. Maybe you thought you only had 20 unread emails, but now you have 500 because all those social updates you ignored are suddenly cluttering your main inbox. After disabling Gemini and losing all those features, Gmail offers you a second chance with a pop-up nag that reenables all those features, including Gemini.

The other Smart Features toggle, one click deeper in the settings, is the Workspace version. This one turns off Gemini in addition to personalized Drive search, copying loyalty cards to Wallet and Calendar event extraction from Gmail. This is supposed to remove Gemini from Drive as well, but you may not see any difference after flipping the switch because the Gemini UI elements don’t go away. Clicking on any of them will produce a prompt to turn Smart Features (and Gemini) back on.

disable Gemini in Gmail

There are two ways to disable Gemini in Gmail, and neither is labeled as such.

Credit: Google

There are two ways to disable Gemini in Gmail, and neither is labeled as such. Credit: Google

So disabling Gemini features means contending with vaguely worded menus and the loss of unrelated features, which looks like a combination of “obstruction” and “forced action” dark patterns. “Clearly, that’s not acceptable,” Potel said. “The fact that it changes your usage parameters when you disable Gemini is obviously meant to prevent you from disabling it.”

Default settings are powerful even without predatory design, but the way defaults are managed can easily veer into a dark place. According to Dr. Harry Brignull, who coined the term “dark pattern” in 2010, companies have long relied on defaults to bring users along for the ride.

Speaking about the tech industry generally, Brignull said that companies are acutely aware of this effect and the advantages it can bestow. “We could talk about this as the pre-selection dark pattern,” he said. “If you want to get people to opt in to something, instead of asking them in a sign-up step explicitly with a link to learn more, you can take that and make it pre-selected. Put that in the settings three to four clicks deep, and that way you may feel you have the means to argue that users are given a choice when, in fact, you know from your own stats that by designing it this way, very few people are even getting to that page.”

Google has billions of users hooked on its products, and that gives it a lot of power to get new features in front of users, since people rarely change the defaults. Google has paid handsomely to set defaults on devices like the iPhone for this reason. The issue came up repeatedly in Google’s recent antitrust cases, and government lawyers used it to paint the company as anticompetitive.

What we’re seeing in both free and paid Google accounts is the power of defaults in the AI era. The default is sharing data for AI training. The default is AI summaries in your email. The default is AI-powered document creation. You can change these settings, but Google has to know most people won’t do that, because the options are hard to find and don’t work as they should.

Welcome to the future.

Photo of Ryan Whitwam

Ryan Whitwam is a senior technology reporter at Ars Technica, covering the ways Google, AI, and mobile technology continue to change the world. Over his 20-year career, he's written for Android Police, ExtremeTech, Wirecutter, NY Times, and more. He has reviewed more phones than most people will ever own. You can follow him on Bluesky, where you will see photos of his dozens of mechanical keyboards.

96 Comments