Apple is finally taking the plunge on AI.

The company on Monday unveiled a suite of new artificial intelligence capabilities that will be available in its newest operating system, including connecting its interactive voice feature Siri with OpenAI’s ChatGPT in a major deal that could supercharge adoption of the fast-developing technology.

Siri, for example, will be able to surface answers from ChatGPT for Apple devices and provide relevant contextual information across several apps, the Cupertino, Calif., tech giant said at its highly anticipated developer conference. The iPhone, Mac and iPad maker’s newest operating system update will also feature AI-augmented improvements in its photo editing and image search capabilities, among other things.

Apple Chief Executive Tim Cook described Apple’s new AI-based functions, dubbed Apple Intelligence, as the next big step for the company, which has been slow to adopt emerging technology that has the potential to change the way people live and work.

“Recent developments in generative intelligence and large language models offer powerful capabilities that provide the opportunity to take the experience of using Apple products to new heights,” Cook said in a keynote address during Apple’s Worldwide Developers Conference, where the company previewed the iOS 18 system and other software updates for products including the Mac and iPad.

The move signals Apple’s wider ambitions in the expanding AI landscape, as technology has progressed dramatically. Tools made by San Francisco-based OpenAI have been used to create music videos, read bedtime stories to children and help brainstorm ideas for writers. Companies including Microsoft and Google have aggressively incorporated AI into their products and services.

Apple has often not been the first to market with new technological advances, choosing instead to enter new product categories — including smartphones and tablets — once they’ve been established, leading to broader consumer adoption. For example, Apple only began selling its own virtual and augmented reality headset (known as Vision Pro) early this year.

Apple said its AI capabilities were created with privacy protections in mind. Apple Intelligence uses on-device processing. For requests that require use of the cloud, iPhone, iPad and Mac “do not talk to a server unless its software has been publicly logged for inspection” and the data are not retained or exposed, the company said.

Apple presented several uses for Apple’s new AI features. For example, if an iPhone user gets a notification that a work meeting has been moved to a later time, she can ask Siri how much time it would take for her to get from where the meeting is located to her kid’s play that night. In another hypothetical instance, an iPad user could share a photo of an empty patio and ask Siri what plants should be added.

The company also said customers can use Apple Intelligence to make suggestions for their writing, using it to analyze the tone of an email with options to make it more friendly or professional.

The announcement of the OpenAI deal “kicks off a new frontier for Apple,” said Daniel Ives, a managing director at Wedbush Securities who follows Apple.

“This was a historical day for Apple and Cook & Co. did not disappoint in our view,” Ives, who has an “outperform,” or “buy,” rating on the company’s stock, said in a note to clients. “Apple is taking the right path to implement AI across its ecosystem while laying out the foundation for the company’s multi-year AI strategy across the strongest installed base of 2.2 billion iOS devices over the coming years.”

Investors were less impressed, sending Apple’s stock down 1.9% to $193.12 a share.

Apple hopes adding new AI tools to its products and services will make them more useful to customers and thus more attractive. The company has faced a number of challenges, including slowing device sales in China. Ives said that AI technology introduced to Apple’s ecosystem will bring more opportunities for Apple to generate revenue.

Through its deal with OpenAI, Apple’s digital assistant Siri can ask Apple users if Siri can relay a question to ChatGPT for further information. This allows Apple to harness ChatGPT’s platform and in return, Apple users also become familiar with ChatGPT and what it can do. Every day, Apple said, Siri gets 1.5 billion voice requests.

ChatGPT will be available for free to Apple users on its newest operating systems for iPhones, iPads and Macs later this year. Apple said its users won’t need an account with ChatGPT to use it on Apple devices. OpenAI won’t store requests and IP addresses will be obscured, the company said.

“Together with Apple, we’re making it easier for people to benefit from what AI can offer,” OpenAI CEO Sam Altman said in a statement.

Some tech companies, including Apple, didn’t anticipate the breakthroughs in AI over the last year, said Rob Enderle, principal analyst with advisory services firm Enderle Group. The partnership with OpenAI is one way for Apple to catch up. One of OpenAI’s major backers is Microsoft, an Apple competitor.

“Apple’s been significantly behind on AI,” Enderle said. “This is a method to allow Apple to make up for the fact that they haven’t been focused on AI like they should have done over the last decade or so.”

Apple Intelligence was one of many announcements and updates from Apple on Monday, including a feature that lets AirPods Pro users nod yes or shake their heads no to Siri’s questions when they are in crowded spaces. Additionally, the company announced that the Vision Pro headset will also be available in additional countries starting later this month, including mainland China, Hong Kong, Japan and Singapore.

The company also unveiled a new feature called InSight for its tvOS18 that is similar to Amazon’s X-Ray and shows the names of actors or a song playing on an Apple TV+ program.

OpenAI has become the best-known player in the artificial intelligence space, thanks to its tools including ChatGPT and Sora, its text-to-video tool. But the company has faced its fair share of controversies and challenges.

OpenAI last month received backlash from actor Scarlett Johansson, who said she was approached by the startup’s CEO to record her voice for a Siri-like voice assistant version of ChatGPT. After she declined the opportunity, Johansson said, she was upset when she heard what sounded like her voice in a ChatGPT demo.

Altman is known to be a fan of the 2013 movie “Her,” in which Johansson plays “Samantha,” the disembodied voice of a computer who provides friendship and, eventually, love to a lonely man played by Joaquin Phoenix.

OpenAI said that the AI voice, called “Sky,” was not Johansson’s and was recorded by an unnamed voice actor. Nonetheless, the company paused the use of the Sky voice.

OpenAI recently caught flak for disbanding a team that was tasked with coming up with systems to prevent the rise of artificial intelligence from leading to disaster for humanity. After the firestorm, OpenAI created a new safety committee led by board members, including Altman.

Last week in an open letter, former and current OpenAI employees also raised concerns. The group said that “AI companies have strong financial incentives to avoid effective oversight, and we do not believe bespoke structures of corporate governance are sufficient to change this.”

OpenAI said in a statement said that it believes “rigorous debate is crucial” and it will continue to engage with communities, governments and civil society. The company said it has an anonymous hotline and a safety and security committee.

“We’re proud of our track record providing the most capable and safest AI systems and believe in our scientific approach to addressing risk,” the company said.

Large tech companies are also facing their own challenges, with the U.S. government raising antitrust concerns.

In March, the Department of Justice sued Apple, accusing the tech giant of stifling competition and leveraging its clout and ownership of the popular App Store to increase prices for customers. Apple said the lawsuit threatens “who we are.”

“If successful, it would hinder our ability to create the kind of technology people expect from Apple — where hardware, software, and services intersect,” Apple said in a statement.

Ozempic, Wegovy and other new weight-loss drugs have proved so good at helping users shed pounds, they’ve quickly become a multibillion-dollar industry.

The prescription-only medications have also been in consistently short supply, which is why they’ve grown increasingly popular — with scammers.

Online con artists are luring victims with discount offers of Ozempic and similar drugs with no prescription required. After they take the money, however, they deliver something their clients didn’t order — fake drugs, perhaps, or just the disappointment that comes when people realize they’ve been taken.

A new report by threat researchers at McAfee found 176,871 phishing emails and 449 malicious websites tied to offers of Ozempic, Wegovy and semaglutide, the generic name for these drugs, from January to April 2024. Phishing attempts were almost 200% higher during the period than they were from October to December, the internet security company reported.

In addition, the researchers found that scammers were creating fake profiles on Facebook so they could run weight-loss-drug swindles there. Others took hundreds of fake offers to Craigslist — including 207 of them in a single day in April.

Novo Nordisk originally developed the semaglutide it dubbed Ozempic as a treatment for Type 2 diabetes, but clinicians found that semaglutide could help people lose significant amounts of weight by suppressing appetite. The Food and Drug Administration approved Novo Nordisk’s Wegovy as a weight-loss drug in 2021; since then, it has approved an alternative drug, Eli Lilly’s Zepbound, which is based on its diabetes treatment Mounjaro.

Although Ozempic costs nearly $1,000 a month without insurance, the demand for these drugs has grown rapidly. Sales of Ozempic alone are projected to reach $11 billion this year, according to one analysis.

The combination of high prices and insufficient supply has proved irresistible to scammers.

Abhishek Karnik, head of threat research at McAfee, said the fraudsters typically have two types of victims: people who can’t get a prescription for Ozempic, and people who have a prescription but can’t find it at their local pharmacies.

The scams can be personalized and targeted at people who’ve shown some interest in weight-loss drugs, using information collected about them and their browsing habits, said Iskander Sanchez-Rola, director of privacy innovation for the internet security company Norton. The pitches may come through email or ads placed on search engines or websites, he said, including sites that are well-established and trustworthy.

“Anywhere a human can have their eyes on, they will be there,” Sanchez-Rola said of the scammers. Just because a website is legitimate, he added, that’s no guarantee that the ads there will be.

To pull off the scam, Karnik said, the fraudsters will often interact with the prospective buyer through a social media network or platform such as Telegram to win their trust. That could include offering over-the-top testimonials to their legitimacy and to the quality of the products. “You’ll have people claiming they had huge success with these drugs,” he said, “but none of it is true.”

Scam sellers may also pose as doctors or pharmacists, often from foreign countries, and claim they can sell Ozempic without having to examine you or see a prescription. That may seem sketchy, but many Americans have imported real medications such as insulin illicitly from Canada and Mexico for years because the prices are so much lower outside the U.S.

“One example on Facebook Marketplace included a ‘Doctor Melissa’ based in Canada who could provide Mounjaro and Ozempic without a prescription, with payment available through bitcoin, Zelle, Venmo and Cash App — all of which are nonstandard payment methods for prescription drugs and should be red flags for consumers,” McAfee said.

According to McAfee, some scammers just take your money and disappear, possibly after getting you to share sensitive personal information (unwittingly, in many cases). Others will deliver an injection pen — the typical format for these weight-loss drugs — filled with something other than the advertised medication; they may be insulin injectors, EpiPens or even injectors loaded with salt water, McAfee said.

That sort of counterfeit shipment poses a significant health risk. For example, McAfee said, one person who used Ozempic to help manage her diabetes bought some injectors online after local pharmacies ran out, only to discover that the pens she received were filled with insulin. Had she not been tipped off by the flimsy packaging and different appearance, McAfee said, she could have injected herself with a fatal dose.

Another type of con, Sanchez-Rola said, is when the scammer will deliver a bottle of aspirin or some other drug you didn’t order, then make it so burdensome for you to obtain a refund that you give up.

How to detect Ozempic scams

The first rule, McAfee said, is never to buy one of these drugs without a prescription. After all, doing so is illegal in the United States.

Sticking to licensed pharmacies is wise too. You can check whether a California pharmacy is licensed at the State Board of Pharmacy website; for other states, consult the FDA’s website.

But scammers also target people who have prescriptions they can’t fill locally, as well as offering medications they tout as nonprescription alternatives that are just as good as Ozempic. And to make their products more attractive, they may use AI tools to produce eye-popping before-and-after images that are persuasively realistic.

Here are more red flags to look for before buying a weight-loss drug online:

Strikingly deep discounts. Fraud experts say that if a price looks too good to be true, it almost certainly is. Another thing to bear in mind, Sanchez-Rola said: “You didn’t find the best deal, the best deal found you, which is already a big red flag.”

Misleading claims. McAfee warns that overly rosy promises of results are a sign of a scam. Be especially wary if the site offers none of the usual disclaimers about side effects, possible negative reactions or details about how the product should be used.

Payment methods other than credit cards. Scammers prefer systems that act more like cash, such as Zelle, Cash App or gift cards, or are untraceable, such as cryptocurrency. Sanchez-Rola said sometimes scammers will also offer a credit card option that looks real, but it’s designed to display an error message when you try to use it so you’ll be forced to use a different, sketchier payment method.

A mix of 5-star and 1-star reviews. Sanchez-Rola said that fraudsters’ websites often try to bury the actual reviews posted by unhappy customers under a slew of bot-generated praise. If you see a lot of 5-star reviews that were posted within a short period of time, that’s a huge red flag, he said, especially if the reviews have no comments attached.

Deep discounts that expire soon. Con artists will try to override your reservations about a transaction by giving it a false sense of urgency.

Boilerplate company information. Scammers’ websites often provide phone numbers, addresses, contact information and descriptions that they copy from legitimate sites, Sanchez-Rola said. You should paste the phone number and other information into a Google search to see if they’re used by other, unrelated businesses — for example, he said, one scam site copied its physical address from an ice cream parlor, assuming that its customers wouldn’t bother to check.

Use security software that helps detect scams. McAfee and Norton, among other companies, offer programs that can alert you when you’re about to navigate to a suspicious website.

What to do if you’ve fallen for an Ozempic scam

If you’re fortunate enough to have used a credit card, you can dispute the charge and eventually obtain a refund. You can get similar results if you make your purchase using PayPal or Venmo with the buyer protection feature enabled.

If not — for example, if you used Zelle or paid with gift cards — you can at least report the fraud to try to protect other potential victims. The federal government has an online tool to help you find the right law enforcement agency to file your report with. You can also file a complaint with the FTC’s site and the FBI’s Internet Crime Complaint Center.

Beyond that, Sanchez-Rola said, if you were conned on a social network, you should report the fraudster’s profile to the network’s administrators. For example, Facebook explains how to report fraudulent Marketplace sellers in its help section, and TikTok walks through how to report a problematic account in its support section.

The Supreme Court on Monday said it is putting off a ruling for now on whether social media laws adopted by Florida and Texas violate the 1st Amendment.

Instead, the justices sent those cases back to lower courts to consider how those laws would apply in specific situations.

Speaking for the unanimous court, Justice Elena Kagan said the lawyers for NetChoice, the group that sued the states, and the lower court judges who ruled so far made a mistake by focusing broadly on free-speech principles without considering how the laws would apply in different circumstances.

“In sum, there is much work to do below on both these cases, given the facial nature of NetChoice’s challenges. But that work must be done consistent with the 1st Amendment, which does not go on leave when social media are involved,” she said.

All nine justices agreed with the outcome.

Monday’s decision leaves unresolved whether states may play a role in deciding what appears on popular platforms that are seen by tens of millions of viewers.

The two largest red states had passed laws to fine and punish platforms like Facebook, YouTube, Twitter (now X) and Instagram for what they said was “censoring” posts that appeal to conservatives.

The Florida and Texas laws under review arose from complaints three years ago that President Trump had been discriminated against or unfairly blocked by social media sites, including Twitter.

In 2021, Florida Gov. Ron DeSantis signed his state’s first-in-the-nation law and said it targeted the “Big Tech censors” who “discriminate in favor of the dominant Silicon Valley ideology.”

The measure, adopted before billionaire Elon Musk purchased Twitter and changed its name to X, applies to social media sites with more than $100 million in annual revenue or more than 100 million users.

It authorizes lawsuits for damages for “unfair censorship” and large fines if a social media site “deplatforms” a candidate for office.

Texas Gov. Greg Abbott signed a somewhat broader bill a few months later, saying “conservative speech” was being threatened. It says a social media platform with more than 50 million users in the United States “may not censor … or otherwise discriminate against expression” of users based on their viewpoint.

NetChoice and the Computer & Communications Industry Assn. sued to challenge both laws on free-speech grounds, and both laws were put on hold, including by a 5-4 order from the Supreme Court.

The drive to restrict social media is heating up in many states.

Last week, the court in a 6-3 vote threw out a lawsuit brought by Republican state attorneys that accused the Biden administration of censoring social media.

The administration said it had merely alerted sites about dangerous disinformation about vaccines and COVID-19. Justice Amy Coney Barrett said the state attorneys did not show that Facebook and other social media platforms removed postings because they were pressured to do so by the government.

Last year, the California Legislature adopted a measure to prohibit online companies from collecting and selling data on children and teenagers, but it was blocked on 1st Amendment grounds by a federal judge in San Jose.