Deeplinks

With the launch of iOS 27, Apple is rolling out a variety of new AI features to its familiar voice assistant, Siri. But how are AI tools like these handling our data? In our latest EFFector newsletter, we're talking about the privacy complications of AI phone features.

JOIN OUR NEWSLETTER

For over 35 years, EFFector has been your guide to understanding the intersection of technology, civil liberties, and the law. This issue covers drones and our right to record the law enforcement, video doorbell footage privacy, and how to limit what data Apple's new Siri AI can access.

Prefer to listen in? EFFector is now available on all major podcast platforms. This time we're asking EFF's Thorin Klosowski about the difference between AI phone features that are computed on-device and ones that are computed on external servers—and what that means for data protection. You can find the episode and subscribe on your podcast platform of choice:

Want to protect your right to digital privacy? Sign up for EFF's EFFector newsletter for updates, ways to take action, and new merch drops. You can also fuel the fight for privacy and free speech online when you support EFF today!

San Francisco's decision to retain its use of Automated License Plate Reader (ALPR) surveillance cameras belies what we know about this spying technology tool: it endangers residents and threatens the privacy and civil liberties of our community. Based on what's in the city's press release and announcement, this policy will do nothing to stop actual harms.

We know innocent drivers will be stopped and menaced by officers because of erroneous matches. We know officers use Flock to stalk potential and past romantic partners. Data will be accessed by Immigration and Customs Enforcement (ICE) and used to deport immigrants. Promising greater penalties for such abuse will not end this. These are not isolated mistakes that another policy can fix. They are consequences of building a system that records everyone’s movements and makes them searchable by police. This is also not unique to a single vendor. From Flock Safety to Motorola to Axon—San Francisco must end its use of ALPRs.

We ultimately cannot rely on new protocols from city officials, and the City’s new policy is woefully inadequate. There is no warrant requirement to search stored ALPR data. An incident or computer-aided dispatch (CAD) number is not judicial authorization. Without a warrant requirement, officers can search stored location data without showing probable cause to a judge, and will. Without judicial control, officers will continue to search the data for abusive reasons. But a warrant requirement alone would not justify retaining the ALPR network: the community is demanding an end to the collection itself.

Additionally, the announced policies include no deadline to delete ALPR data, there is only a 30-day deadline to move data from the vendor’s servers to the city’s servers. City officials must understand that moving data is not deleting it. Whether Flock or SFPD stores the data, it remains a permanent record of where people drive, worship, work, organize, seek care, and spend time with others. Thus, the best practice is deletion. New Hampshire requires deletion in three minutes, and Flock itself has reduced the default retention time to seven days. San Francisco can and should do better.

Lastly, while transparency and documentation are important concepts, better audit logs are not the answer. They can expose abuse only after a search has occurred. They cannot undo the disclosure of someone’s movements or justify collecting everyone’s location data in the first place; especially when we are talking about people’s lives and civil liberties. The city's announced policy does not even require officers to state, in their own words, why they are searching the stored ALPR data—an accountability rule that has exposed abusive searches across the country.

San Francisco is behind many communities that have considered the tradeoffs of ALPR surveillance and made the right choice by ending their contracts. San Francisco must do the same.

People are increasingly concerned about the ways in which mass surveillance is tracking our every move: from Flock license plate readers to face recognition to creepy ads that – based on what we see and do online – seem to know everything we’re thinking and planning. It didn’t have to be this way, and since the early days of the internet, a dedicated band of activists, lawyers and technologists have fought for a better, more secure and private digital future – a future that’s still attainable.  

Cindy Cohn, who just finished a 26-year run with the Electronic Frontier Foundation including 11 years as its executive director, has lived this fight. She says privacy isn’t just about secrecy: It's ultimately about power – who has it, and who has the ability to protect themselves from it.   

Welcome to the first episode of “Privacy’s Defenders,” a podcast about the people – lawyers, journalists, hackers, and others – who’ve fought to secure your digital liberties since before most people even knew what the internet was.

(You can also find this episode on the Internet Archive and on YouTube.)

In this episode, Cindy talks with EFF cofounder John Gilmore about how he – an early employee at Sun Microsystems – came together with Lotus Development cofounder Mitch Kapor and cattle rancher, philosopher and Grateful Dead lyricist John Perry Barlow to create EFF as a bulwark against government investigation and prosecution of early internet users.  

It’s a story of the Secret Service’s “Operation Sundevil,” jet-setting tech titans, tie-dyed cypherpunks, and a fateful house party in San Francisco’s Haight-Ashbury district amid the earliest days of online communications, setting the stage for the battles that created the internet as we know it and issues we still grapple with today. 

The “Privacy’s Defenders” podcast is a follow-up to Cindy’s book, “Privacy’s Defender: My Thirty-Year Fight Against Digital Surveillance,” bringing to life pivotal moments in the voices of those who fought for your rights. Sales of “Privacy’s Defender” benefit EFF, so pick up your copy today! 

Joanne Elgart Jennings co-produced and created this podcast.  

Jarrod Sport co-produced, mixed, and mastered it.  

Corinne Ruff is our story editor.

We had additional help from Rachel Estabrook and Alison Broverman.

The original music was composed and performed by Nat Keefe of Hot Buttered Rum with Ben Andrews on the fiddle.  

And other archival sound came from the Internet Archive's amazing collection, including the snippet of the Grateful Dead song “Cassidy” that John Perry Barlow co-wrote. 

The San Francisco Police Department (SFPD) began regularly deploying drones two years ago and has since expanded their use in a way that has outpaced its documented policy and evaded existing local and state oversight of these devices. 

The department has a new proposed policy, which continues to be grossly inadequate in protecting privacy and civil liberties. At best, the draft policy continues the SFPD’s pattern of putting vague guardrails on a powerful surveillance tool, but at worst, if implemented, the policy could effectively usher in sweeping, non-targeted, and unspecified general surveillance over the city with few guardrails.

EFF has repeatedly opposed the unaccountable development of the SFPD’s drone program and recently sent a comment to the Police Commission, the local civilian oversight body, about the SFPD’s new proposed policy. 

The SFPD has been sidestepping oversight of its drones since 2024. In March 2024, San Francisco voters approved a heavily-funded, billionaire-backed measure, Proposition E, which sought to expand police access to surveillance technology. Among its impacts, Prop E removed drones from oversight required by the 2019 Surveillance Technology Ordinance. Nonetheless, in its haste to purchase drones after Prop E passed, the SFPD knowingly violated California’s AB 481, a state statute requiring law enforcement agencies to get approval from their local elected governing body before purchasing military equipment, including drones. Eventually the SFPD sought retroactive approval from the Board of Supervisors and, soon after, announced that it would be launching a drone-as-first-responder (DFR) program.

Now, San Francisco finally has an opportunity to update the SFPD’s guidance in a way that won’t quickly become stale, as has happened while the SFPD steadily increases the purposes for drone use. Though drones were initially identified as tools to use for specific actions such as vehicle pursuits and active criminal investigations, within a year, the SFPD expanded use cases to include patrol, i.e. unrelated to a specific incident. Along with this mission creep, the SFPD has also steadily and exponentially increased the number of drone flights, from roughly 350 deployments in 2024, to over 1,100 from January to August 2025, to over 3,500 in just the first five months of 2026.

The original draft of an updated policy brought by the SFPD to the local Police Commission, a civilian oversight body, earlier this month provided limited details and proposed allowing police to treat drone flights as an extension of their patrol abilities, paving the way for general surveillance, including of First Amendment-protected activity. The proposal received significant community pushback, and the San Francisco Public Defender’s Office authored a letter describing the policy’s shortcomings. That letter was signed by over a dozen local, state, and national groups, including EFF. 

Based on these concerns, the Police Commission deferred taking action until the SFPD addressed them. The SFPD then revised its proposed policy, but this, too, falls short of providing practical guidance to officers and protecting civil liberties, as the Public Defender’s Office identified in a follow-up letter signed by over 40 organizations, including EFF.

EFF’s additional comment to the Police Commission, in part, calls out the incredible gap in oversight of these ballooning drone flights and the immense data collection they facilitate:

The revised policy states that “[unmanned aerial vehicles] may be used as an asset in any situation in which a member may be deployed for a public safety response or when a member onviews criminal activity” but fails to define what is meant by a “public safety response.” The revised policy also provides a definition of “Drone as First Responders,” but it fails to provide any more detail about appropriate DFR deployment. Without appropriate safeguards around deployment and use, drones could be deployed to every call for service, even in situations that are ultimately deemed nonincidents, collecting data along the way that is then stored for 30 days. This type of general patrol could effectively become general surveillance, which SFPD acknowledges is an inappropriate use of their drones and yet is still possible under the vague terms of the current DGO. 

The Police Commission is set to consider the matter on October 14. You can read EFF’s full comment here.

EFF legal intern Simar Kaur also contributed to this article.

Americans’ First Amendment right to equal access to official government statements is violated by the Trump administration’s use of Truth Social’s preferential treatment scheme, which blocks people who won’t pay Trump’s company up to $100,000 a month for early access to government news, EFF told a federal court.

The First Amendment guarantees that members of the public have equal access to public officials’ public comments, we reminded the court.

EFF filed an amicus brief in support of a motion for a preliminary injunction in the lawsuit filed by The Intercept Media and the Freedom of the Press Foundation against President Trump and other administration officials. The lawsuit challenges their use of Truth Social as their primary social media method of making official announcements when that platform provides people who pay a fee for early access to such posts. 

Trump uses his Truth Social account as his primary means of communicating with the public, including to announce military operations and ceasefires, foreign and domestic policy, and the removal and appointment of heads of federal agencies. Earlier in the year, Trump Media, which owns Truth Social, announced “Truth API,” a service that provides investors early access to “market-moving” messages from the president and other high-ranking officials for a fee of up to $100,000 per month.

The plaintiffs, the Freedom of the Press Foundation and The Intercept, contend that the president and other officials’ preferred use of Truth Social with this service violates the First and Fifth Amendments of the Constitution. The plaintiffs are asking the court to immediately prevent the president from posting on Truth Social in a manner that allows him to profit from selling early access to government information.

EFF’s amicus makes two main points.

First, the brief establishes that social media is pervasively used by government officials and agencies as a medium for official communication with the public, including to disseminate critical public safety information and make official announcements.

Second, the brief explains that the challenged practice violates the First Amendment, which guarantees a right to access public officials’ public comments on equal terms with other members of the press and public. Giving some people preferential access must at a minimum be reasonably justified to satisfy First Amendment scrutiny, a test the administration does not meet.

Lining the president and his company's pockets is not a legitimate government interest for restricting timely access to the government's statements. Further, the fact that the public could ultimately access the information from other, less direct channels does not eliminate the need for First Amendment scrutiny; mere delays in timely access still trigger First Amendment scrutiny.

EFF has been advancing the First Amendment right of equal access to government’s public social media posts since at least 2018. We’ve argued that the right of equal access, which is well established in offline contexts, must apply to official government social media posts as well. This case presents an excellent opportunity for a court to directly adopt that position. 

Online sports betting company DraftKings is using AI to target customers who are most likely to place losing bets and respond to gambling promotions. This kind of targeting is a form of online behavioral advertising, which is when companies personalize the ads they show you based on the data they’ve collected about you. The more data a company has, the more personalized the ad can be. While DraftKings is using AI to supercharge the harmful effects of online behavioral advertising, EFF has long argued that all behavioral advertising should be banned. 

According to the New York Times, DraftKings is using its customers’ betting records to train a machine learning model to find losing gamblers. Once found, DraftKings sends these customers targeted advertising designed to lure them back to the site to place more bets—bets that DraftKings thinks will be losing ones. DraftKings has a business incentive to keep losing gamblers coming back to their site, because these are the users actually making DraftKings money. Unfortunately, those considered “problem gamblers” (people who repeatedly gamble despite harm to themselves, their finances, and their relationships) are highly likely to be targeted by this model. By re-engaging these individuals through targeted promotions aimed at keeping them on the platform, DraftKings is capitalizing on their vulnerability for profit instead of mitigating their risk. 

Predatory online behavioral advertising isn’t new, but companies’ use of AI to process data and target customers has magnified its harms. Online behavioral advertising incentivizes the collection of vast quantities of data to power ad tech. Adding AI into the mix means that even more data is collected to train and refine models. Because AI operates as a black box, the humans building the models can rarely predict which data points are the most useful to the AI, driving them to continuously collect more data. AI also allows companies to process enormous data sets much faster, and, as a result, supercharges the harms of online behavioral advertising. 

A direct consequence of online behavioral advertising is that it provides the data the surveillance industry needs to run. Data collected for targeted placement of ads is being sold to insurance companies, banks, and state and federal government law enforcement agencies such as CBP. ICE is also taking an interest in the data fueling ad tech: earlier this year, ICE published a Request for Information “seeking information to better understand how the industry’s commercial Big Data and Ad Tech providers can directly support investigations activities.”  

DraftKings seems to be using solely “first party data” to target their ads, meaning that they’re using only the data they collect directly from their users and are not buying any additional data from third parties to fuel their machine learning model. This highlights how policy solutions that only limit third-party data sharing and selling would not be enough to prevent these predatory advertisements. Rather, policymakers must ban online behavioral ads.   

What DraftKings is doing with their targeted promotions is just one example of how online behavioral advertising causes real harm to real people. But there are ways to take back control over your own data: EFF offers resources such as our Surveillance Self Defense project, along with other tips for how you can protect yourself on mobile apps and on websites.  

DraftKings’ use of AI to target losing gamblers illustrates how ad tech evolves and how companies find new ways to use our data against us. This is why EFF believes that all behavioral advertising should be banned. If companies can’t send personalized ads, they’ll have less incentive to collect the behavioral data powering them.  

EFF joined an amicus brief with ACLU, ACLU of D.C., National Press Photographers Association, and Professional Photographers of America to urge the D.C. Circuit to vacate an FAA drone flight restriction that violated the First Amendment right to record law enforcement. This is an important case—Levine v. FAA—challenging the ability of the government to punish drone pilots who record law enforcement officers engaged in official business. 

As we wrote about earlier this year, the FAA issued a flight restriction for drones that had effectively criminalized the recording of Department of Homeland Security officers, including immigration agents from ICE and CBP, and their vehicles (what the FAA called “mobile assets” including “ground vehicle convoys and their associated escorts”) even if the drone was over half a mile away. 

A drone operator, represented by the Reporters Committee for Freedom of the Press, sued the FAA in March [PDF]. But in April, the FAA rescinded the flight restriction. 

The petitioner argued in his opening brief that the court should evaluate the legality of the flight restriction even though it was withdrawn. Drone pilots could still be punished for violations that occurred when the flight restriction was in effect. And the FAA could reinstate the flight restriction at any time, given that the rescission did not seem to reflect “a true change of heart” but rather an effort by the agency to avoid judicial review. 

The amicus brief, filed in support of the petitioner, noted that drones are unique because they provide “perspectives that cannot be captured by ground-based imagery,” and they “are far more maneuverable than ground-level cameras, and they are both much cheaper and much safer than using a chartered plane or helicopter to record newsworthy events from above.” The brief highlighted that drones have captured “bird’s-eye images of protest activity” and “police uses of force against protestors,” and have “allowed journalists to provide the public with up-to-the-minute information about natural disasters without putting themselves in harm’s way.” 

The brief argued that using drones to capture images and video is information-gathering activity protected by the First Amendment (similar to using cell phones to record law enforcement). The brief also argued that the FAA’s flight restriction appeared to be issued specifically to ban the recording of immigration agents and thus hinder accountability for their enforcement actions—it surely wasn’t a coincidence that the FAA imposed “no-drone zones around all roving DHS patrols just as those patrols were provoking intense national backlash.” If that’s true, it would make the FAA’s action a content-based restriction on speech that is subject to strict scrutiny—the highest First Amendment standard—and presumptively unconstitutional. And even under less rigorous standards of First Amendment scrutiny, the flight restriction is unconstitutional because the FAA can’t articulate any valid governmental interest justifying such a sweeping restriction on speech. 

Resolving this issue to protect First Amendment rights is especially urgent as government agencies continue to sink billions of dollars into technology designed to counter drones—technology that could easily be deployed against journalists and other people hoping to use drones to document government abuse.  

We urge the D.C. Circuit to review the petition and to vacate the FAA’s flight restriction, which would send a message that the government can’t avoid accountability by punishing those who exercise their First Amendment rights. 

The EU Commission draft law to restrict young people’s access to the internet that it presented last week will come at a high cost: it will put online services behind age gates, expand the use of intrusive age verification, and undermine the privacy of all users. 

The EU Kids Act aims to protect children from risks associated with social media, video games, and AI systems by introducing age-based access rules, safety requirements, and stronger enforcement and oversight measures. It presents itself as building on the Digital Services Act (DSA) and puts into “hard law” some of the safety-by-design measures specified in the non-binding DSA guidelines on minors’ protection. 

The proposal is built around the following elements: social media age “delay”, safety by design, age assurance and parental responsibility, and strong enforcement. Each of these measures are concerning.  

Following the advice of an expert panel, the proposal would create a phased access to social media and video-sharing platforms deemed risky—a threshold met simply by relying on personalized recommender systems or offering “uninterrupted content consumption”: no service accounts for children under 13; restricted accounts under tight parental supervision from 13 to 15; and autonomous accounts in a safe-by-design environment from 15 to 18. Full online access is therefore reserved for adults. 

However they’re designed, age gates undermine civil liberties, reduce safety, and create barriers to internet entry, often at the expense of marginalized groups.

If this sounds complex and like a compliance nightmare, that’s because it is. The access delay comes with privacy-intrusive age verification across the board, relying on the EU age verification scheme. For teenagers, this law means significant control in the hands of their parents, who must set up accounts and prove that they are, in fact, parents, adding yet another problematic layer of verification. 

In fairness, the Kids Act’s gradual approach at least appears to be designed with some proportionality considerations, rather than imposing a blanket social media ban. Just last month a French court declared such undifferentiated bans unconstitutional. The EU Kids Act distinguishes between age groups and certain services and follows a risk-based approach. This means, for example, that age verification is not required for existing accounts if the provider can tell with a “high degree of confidence” that the user is above the age threshold—a vaguely specified standard.  

Yet, the law still indiscriminately covers social media and video-sharing, with virtually all mainstream services being covered by the proposal. The broad scope also sits uneasy with the use of age thresholds, which remain a blunt proxy for maturity. What is more, by focusing heavily on safety and harms, the EU Kids Act pays little attention to the privacy and freedom of expression rights of users, as well as the right of children themselves to access information and to participate online. However they’re designed, age gates undermine civil liberties, reduce safety, and create barriers to internet entry, often at the expense of marginalized groups. They also create a powerful infrastructure for control and further entrench the power of big tech. 

The proposal exempts not-for-profit encyclopedias, scientific repositories and educational services, as well as open-source software-developing and-sharing platforms. However, no exceptions are foreseen for small and medium-sized enterprises, which will only foster the dominance of resource-laden tech companies that were already investing in similar measures. And we know that most companies are well-advised to play it safe and use privacy-unfriendly age checks across their platforms. 

The proposal’s second pillar, “safety by design”, casts a wider net. It applies across social media, video-sharing, online games, AI companions, chatbots and even app stores—with varying requirements. Providers must generally make child-safe design the default and can relax from the requirements only if they use age assurance to establish that the user is an adult. 

For example, rules on addictive features such as infinite scrolling, safe account settings, and more choice over recommender systems are to provide a safe internet experience to young people. As regards AI companions and chatbots, the proposal requires companies to design their services to reduce minors’ exposure to emotional dependencies and harmful interactions. Online games are covered as well: they must come with contact protections. The law also makes app stores the gate keeper for age-appropriate access, based on an age-rating system. 

The devil of these measures lies in the details, but all of them raise fundamental rights concerns and some of them seem poorly suited, if at all, to the decentralized architecture of the Fediverse. The requirement for very large online platforms to set up compliance plans before rolling out new services raises additional questions about the risks of transplanting product-safety doctrines of conformity and risk control into speech regulation. Deciding what is “safe” can easily become a question of what content people can access or share.  

By choosing to regulate all these aspects through the Kids Act, the Commission not only but creates a privacy minefield, it also intermingles the digital fairness agenda with the more fundamental-rights heavy questions of age assurance and access to information. An unfortunate policy choice that will politicize well-intentioned efforts to curb manipulative and addictive design practices (read our position on the DFA). 

It speaks volume that the Kids Act has not gone through a full impact assessment process, which would typically require a systemic check of alternative policy options and stakeholder consultations. Looking forward, we call on the EU lawmakers to pull the teeth of the most harmful suggestions and to make sure that the new measures don’t erode the fundamental rights of all users. 

California Gov. Gavin Newsom's executive order is an opportunity for a needed, thoughtful conversation about artificial intelligence and its potential harms. Everyday Californians are feeling real anxiety about the risks of artificial intelligence, and as an organization that works to ensure technology empowers people, EFF welcomes this order as a way for the state of California to lead a much-needed dialogue that addresses these concerns. 

DONATE TO EFF

ensure ai serves people, not power

Nonetheless, the most immediate and current concerns with this technology are not about sci-fi scenarios concerning rogue super-intelligence. They are happening right now through biased algorithmic decision-making for employment or government benefits, AI-powered surveillance systems such as Flock cameras, and artificially inflated personalized pricing. People want state and federal leaders to act, and we urge Gov. Newsom to develop thoughtful policies to address those concerns. Today’s EO is a good start. 

To that end, EFF supports the focus on expanding the reporting requirements under SB 53 (2025) for loss-of-control incidents, alongside third-party investigations. We urge the administration to consider how to make these third-party investigations available for smaller developers. As the Government Operations Agency prepares its recommendations for the governor, we urge leaders to also realize that the effectiveness of kill switches in advanced AI systems remains an area of active research. As such, they should ensure that - as we’ve previously mentioned - any technology regulation targeting cybersecurity practices at AI labs must be careful, precise, and practical. Moreover, we also caution that government-controlled kill switches run the risk of being used as a form of retaliation against protected speech, as demonstrated by the Trump Administration’s retaliatory actions against Anthropic earlier this year.

Ultimately, true safety requires California to focus on concrete, immediate, and urgent harms of AI technologies by ensuring that algorithmic decision-making in both the government and private sectors respects people’s rights and well-being. We urge Gov. Newsom and the state of California to develop thoughtful policy in collaboration with those most at risk of harm to address these and other concerns.

Apple’s new operating system is here, and along with it comes a new version of Siri, dubbed with two very familiar letters: AI. As the name suggests, this Siri power-up resembles an AI chatbot more than the often derided voice assistant you might be used to. It has even evolved from a blob you invoke with a verbal command or a button press to a whole app. This update comes with a slew of privacy complications, but you can take some control over what this new Siri can access and use. 

There’s no denying that the new version of Siri is far more powerful than it used to be, and arguably more useful at surfacing details on your phone. But that comes at the cost of deeper access. Once enabled, Siri and Spotlight are combined, unifying the interface. Where you may have once just pulled down on the screen to search for an app or contact, you’re now also invoking Siri. 

Spotlight and Siri are now visually one and the same.

By default, ask Siri a question and it’ll search through your Apple apps, like Notes, Messages, emails, and more. As time goes on, if the developer chooses to let it, Siri will gain access to more and more third-party apps. If an app developer doesn’t add that support, then Siri AI won’t be able to access the contents of that app (unless it’s shared screenshot-style via a new feature called “on-screen awareness,” which we’ll talk about more in a moment). 

For example, if Signal doesn’t choose to implement Siri AI support and you only talk to Bill on Signal, you won’t get an answer when you ask Siri AI, “What was the last photo Bill sent me?” But if you talk to Bill on Apple Messages and ask that same question, Siri AI will summarize what it thinks the photo is.

Sometimes Siri processes this data on your device. Sometimes it uses Apple’s Private Cloud Compute (PCC), which means the data is sent off your device to a cloud server. While you can try digging through the Apple Intelligence Report to figure out what’s sent to PCC, there’s no immediate visual indication from the user’s point of view when data leaves the device or when AI can handle it on the phone, iPad, or Mac itself. In practice, ask Siri AI a question and you’ll never really know if it’s being computed on device or off.

Apple claims what’s sent to PCC is not stored by the company after it is processed, but there are certain types of data or certain apps you might have on your phone that are not worth the risk. That’s especially true if you’re using a feature like Advanced Data Protection, which turns on end-to-end encryption for much of what’s stored in iCloud. Sending data that’s stored with end-to-end encryption off your device and into the cloud—no matter the privacy promises—is a fundamental change to the risk assessment you should make. “Private” means the system is engineered so that Apple shouldn’t be able to see or store the data, but it doesn’t mean it’s encrypted or doesn’t leave the device.

This leaves the privacy of certain apps up to a strange combination of an app developer’s choices and your own. You can, of course, disable Siri entirely (Settings > Siri > "Turn Off Siri"), or choose not to invoke Siri to ask questions, but perhaps you don’t want to fully disable or disengage with the system. Thankfully, you can put some guardrails on Siri AI’s access. Once you’ve updated to iOS 27, here are the steps to take. 

Note: only iPhone 15 Pro/Pro Max, as well as all models of the iPhone 16 and newer support Apple’s AI features. Siri AI is currently only available in English, and not available worldwide.

By default, how (and if) Siri AI can access data inside apps is up to the app developer. If an app developer chooses to index the contents of their app, then it may appear in search, and thus be made available to Siri AI. This means the content may pop up during general or direct searches, like “What are my plans for November" might cull information from your calendar, Messages, Notes, and, as they update, third-party apps.

If you do not want Siri to look through certain apps to consider the contents in results, you can tell it not to:

With this setting disabled, when you ask Siri general questions, it will not surface details from the app you selected. For example, if you disable “Show Content in Search” for Messages, it will not be able to read your Messages conversations. 

Left: Asking Siri to summarize a message thread with Show Content in Search enabled. Right: With the setting disabled.

There is also an “App Access” setting where you can configure some of the ways Siri interacts with apps. You’d think this is where we’d have gone to revoke access to the content of an app, but alas, this settings page is more about some basic functionality with device personalization, not Siri’s access to the contents of the app.

On this screen, you’ll find a variety of options, depending on what an app supports. “Learn from this App” sounds nefarious, but is mostly about tracking usage, like how often you open an app, and if a developer supports it, what you interact with.

The rest of the options are mostly about the personalization tweaks that Siri makes, where it suggests apps it thinks you want at the moment in various places, like when searching or sharing. “Show on Home Screen,” “Suggest App,” and “Suggest Notifications” are just about whether you see apps in those places. 

For example, if you have a widget of Siri-suggested apps on the home screen, that’s the “Show on Home Screen” toggle. If you see an app recommended in another app, like adding a date to your calendar from an email, that’s “Suggest App.” Apple claims these features all use on-device processing and the data is not stored on servers.

For anything not covered here, refer to this documentation for steps to disable certain features.

There is one Siri AI feature you (and app developers) can’t do as much about: on-screen awareness, a feature you can invoke at any point to prompt Siri and ask it to explain what you’re looking at and perform certain actions. For example, you can ask it to summarize a web page, cut a recipe you’re reading in half, add an event to your calendar, or try to figure out where a photo was taken. All potentially useful features.

But you can also ask it to summarize or explain a Signal group chat that you're looking at, or a meme in a WhatsApp chat, and the data from that on-screen interaction may be sent to PCC. There is currently no way for you or app developers to block this feature, so it’s up to you, and those you chat with, to simply not use it if you’re concerned about the content of conversations potentially leaving your device. It would be a large improvement to privacy, especially secure chat apps, if Apple provided developers a means to block access to Siri AI’s on-screen awareness tool. Even better if they gave you a single control to block all Siri AI features from an app entirely.

By default, Siri AI won’t collect and use data from your interactions with it for training AI features. But during the setup process, Apple provides a way to opt in, which you might have tapped without thinking about it. If you’d rather your data not get used for training, you can opt out:

According to Apple’s privacy documentation, disabling this option should revoke training access to the audio and text from the Siri app.

Want nothing to do with any of this but still find Siri useful enough to keep around (or you just have to keep it turned on in order to use CarPlay)? For the time being, you can get the old Siri back, though the process is a bit odd.

You can no longer easily disable Apple Intelligence entirely with one tap in the Settings, but on this screen you can also configure other AI features, like disabling the writing and math assistance prompts, turning off image creation, and disallowing the use of extensions. Follow this guide on Apple's site for everything else.

For the most part, Apple’s handling of AI features is far less in-your-face than others, and because of that the privacy implications are easier to untangle. But even still, it’s difficult to know what’s processed on device and what’s sent off, and so the privacy trade-offs are never spelled out as clearly as they should be. 

Apple could improve on this by offering an on-device only option for Siri AI and providing a clear, single setting toggle to prevent all AI features in a specific app (it looks like Apple is planning a single privacy toggle in a future update. We'll update if and when it does). In general, Siri’s power-up has also made it blurry and difficult to really figure out what sorts of privacy options exist. “Siri” means many things, both on device and off, ranging from “searching the entire internet for an answer” to “setting a timer,” and users have no straightforward ways to wrangle that data to suit their needs. As it stands, it’s a confusing collection of different toggles that never feel exactly right and which many users might struggle to grasp.