‘Deepfakes’ get a rebrand
State of play: Meta launched Vibes, soon followed by OpenAI’s Sora, both offering endless AI-generated short video feeds similar to TikTok. Sora’s unique feature lets users insert their own faces into realistic AI videos called ‘cameos.’ To address misuse concerns, OpenAI requires users to opt in to share their likeness but uses an opt-out system for copyright holders’ intellectual property. This led to controversy when the app quickly showed offensive, fabricated videos of celebrities and historical figures, including disrespectful depictions of Martin Luther King Jr.
Why it matters: Deepfakes have been rebranded as ‘cameos’, normalizing what was once a fringe technology and embedding it into mainstream social platforms. OpenAI argues there are free speech interests in depicting historical figures though it allows authorized representatives or estate owners to request removal of their likeness. However, this approach leaves most individuals, especially those without representation, largely unprotected from image misuse. As AI-generated content floods social feeds, the risks to reputation, identity, and trust accelerate. This is the reality: the rebranding of deepfakes as ‘cameos’ has made synthetic media an everyday part of digital life, while regulation and safeguards struggle to keep pace.
Hear from our cybersecurity experts to learn more about how deepfakes are expanding the threat landscape.
Your phone isn’t listening to you, rather now, it’s just AI
State of Play: Instagram head Adam Mosseri took to social media to once again deny that Meta listens in on users through their device microphones to target ads insisting, “we do not listen to your microphone” as they simply don’t need to. Just days later, Meta announced a new data play: the company will soon use the content of your AI chats to personalize ads across Facebook and Instagram. In other words, while the microphone myth is dismissed, Meta is about to gain access to an even more intimate data source, your direct conversations with its AI. This puts harvesting user intent and interests for ad targeting at a new level.
Why it Matters: While Meta assures users it’s not eavesdropping, it’s rolling out a system that’s arguably even more invasive: using your AI chat data to fuel hyper-targeted advertising. This move blurs the line between private inquiry and public data, raising new privacy questions and trust issues for users. For brands, it offers sharper targeting, but at the cost of heightened scrutiny and the need for greater transparency. As AI intent data replaces passive listening, the real test will be whether platforms can balance innovation with genuine user control and informed consent.
Meta bans political ads in EU amid new rules
State of Play: Starting October 2025, Meta will stop allowing political, electoral, and social issue ads on its platforms across the European Union. This decision comes in response to the EU’s new Transparency and Targeting of Political Advertising (TTPA) rules, which Meta says are too complex and create legal uncertainty. On the other hand, political ads remain allowed in the United States, where such ads are protected as free speech and face far fewer restrictions on content or targeting.
Why it Matters: This highlights a growing divide between Europe and the US in handling political ads online. Europe prioritizes transparency and user protection, even if it limits political speech. The US focuses on free speech, allowing targeted political ads to reach voters freely. For political groups, Meta’s EU ban cuts off an important way to connect with people, which could reduce public debate. It also shows that companies like Meta may leave markets rather than follow strict rules, creating a patchwork of different standards for digital campaigning worldwide.
AI knows your posts better than you do
State of Play: Reddit has filed a major lawsuit against Perplexity AI and several data-scraping firms, claiming they illegally took millions of Reddit posts to train their AI systems. The lawsuit says these companies dodged Reddit’s API rules by pulling content through Google searches instead of using approved, paid data access. Reddit argues this scraping breaks its terms of service and takes advantage of the work and creativity of its users without permission or payment.
Why it Matters: This case highlights a growing fight over who owns the data that powers AI. Reddit’s move shows platforms are no longer willing to let AI companies freely use user content for profit. For everyday users, it raises privacy and fairness questions; your online posts could be used to train AI without you knowing. For AI developers and brands, it’s a warning that how they collect and use data is under new legal and ethical pressure.
Tech giants dodge breakup amid shutdowns
State of Play: The federal government shutdown has caused a split in the progress of major antitrust cases against tech giants. Cases against Google and Meta are moving forward, with judges pushing to finalize rulings soon. In contrast, cases against Amazon and Apple have been paused until federal funding is restored, with judges approving delays for depositions and discovery scheduled through early 2027. The length of the shutdown will determine how much these delays impact trial timelines.
Why it Matters: This situation highlights how political and funding challenges can slow down high-profile efforts to rein in big tech market power. Extended delays risk slowing progress on addressing alleged monopolistic practices, potentially leaving consumers and smaller businesses without needed protection. The eventual outcomes of these cases will play a crucial role in shaping future antitrust enforcement in the technology sector.
Instagram hits 3 billion, as antitrust threat fades
State of Play: Instagram has reached 3 billion monthly active users, further solidifying its central role in Meta’s portfolio. To mark the milestone, Instagram is piloting new features that let users control algorithmic recommendations and filter out unwanted topics from their feeds. This comes as Meta’s high-profile antitrust trial with the FTC has concluded. Despite the government’s efforts to force Meta to divest Instagram and WhatsApp, expert analysis suggests the FTC’s case is unlikely to succeed, with the court expected to reject claims that Meta’s acquisitions were unlawful or anti-competitive.
Why it matters: Instagram’s record growth coincides with a likely win for Meta in court, signaling that its current business structure, and its integration of Instagram, will probably remain intact. For users and brands, this means continuity and stability, even as new feed controls shift how content is discovered. For regulators and competitors, the outcome highlights the challenges of using antitrust law to reshape the social media landscape and suggests that meaningful change may have to come from product innovation, not the courts.
