iPhone Duo, AI on Your Wrist, and OpenAI's Math Milestone
Updated on September 10, 20266 minutes read
Apple just shipped its first foldable phone, raised prices across its entire lineup, and leaned harder into AI than at any event in recent memory. Here is what happened this week and why it matters if you are learning to build, design, or secure the products that come next.
Apple's first foldable: the iPhone Duo
The headline product from Apple's "Surprise and Shine" event is the iPhone Duo, a folding phone that starts at $1,999. It has a 5.4-inch outer screen and a 7.6-inch inner display, an IP68 water resistance rating, Touch ID, two rear cameras, and stylus support, making it the first iPhone to work with a pen. Apple also used AI and 3D-printing techniques during manufacturing, specifically to engineer the hinge mechanism at the precision the design required.
One trade-off that sparked immediate debate is the under-screen front camera. Hiding a lens beneath the display keeps the inner screen uninterrupted, but camera performance suffers compared to a conventional placement. Ars Technica has the full spec rundown on Apple's long-rumored foldable, and CNET explains the image-quality compromises of the under-screen camera.
For anyone studying product design or UX, the Duo is a real-world case study in design trade-offs: every choice that makes a device thinner or cleaner has a cost somewhere else.
Every iPhone just got $100 more expensive
Apple did not stop at the Duo. The iPhone 18 Pro starts at $1,199 and the Pro Max at $1,299, each $100 higher than their predecessors. More surprisingly, Apple also raised prices on older models still in the lineup, including the iPhone 16, 17, and iPhone Air. The Verge frames this as part of an industry-wide trend toward smartphone price hikes, and TechCrunch confirms the older-model increases.
The standard iPhone 18 was not announced at all, which CNET attributes to a deliberate strategy around timing and supply. New CEO John Ternus, making his first keynote appearance in the role, leaned into an argument that echoes something Steve Jobs said decades ago: the iPhone itself is your primary personal computing device, and AI makes that case stronger than ever. TechCrunch connects Ternus's framing to that earlier Jobs-era playbook.
Apple goes big on AI, and raises privacy questions
Apple Intelligence got more stage time than any previous software announcement in recent years. New Siri features, a tie-in with the third-party app Stuff for personal context, and a batch of AI audio tools are all part of the push. CNET reports that Ternus explicitly positioned the iPhone as a personal AI device.
The Apple Watch Series 12 and Ultra 4 got the most eyebrow-raising feature: Siri Recap, which can transcribe and summarize ambient conversations. Apple says raw audio is not stored, and it published a document detailing how the listening features are designed to protect privacy. The Verge has that Apple privacy document explained. Still, TechCrunch argues the feature normalizes the idea of technology that is always listening, a question of consent that is worth thinking through carefully, especially for those going into cybersecurity or product ethics.
Not everything announced is shipping immediately. CNET has a breakdown of which features arrive later, including iPhone Handoff, which will eventually let you share one phone number across two devices, though it will be limited to certain carriers at launch. See CNET's explainer on iPhone Handoff.
OpenAI solves a Millennium Prize math problem
Away from Apple Park, OpenAI announced it had produced a solution to one of mathematics' famous Millennium Prize problems, specifically related to the Navier-Stokes equations. These are equations that describe how fluids move and have stumped mathematicians for generations. The circumstances around the announcement were unusual and created friction in academic circles before the result was even formally published. The Verge covers both the achievement and the complicated reaction from academia.
For anyone learning data science or AI, this is worth following: it points to how large AI systems are starting to contribute to formal mathematical reasoning, not just pattern-matching on existing answers.
Separately, OpenAI added Paul Christiano to the board of its nonprofit foundation. Christiano is a well-known researcher in the field of AI alignment, the work of making sure AI systems do what humans actually intend. TechCrunch explains who Christiano is and why his appointment matters.
Suno's AI music model gets record-label backing
AI music startup Suno released its v6 model, and this version is notably different from its predecessors: it was trained on licensed data provided with record industry cooperation, rather than the datasets that led to prior legal disputes. The Verge reports on Suno v6 and what the label partnership means in practice. For anyone interested in creative AI or the legal side of machine learning, the shift from unlicensed to licensed training data is a meaningful development in how the industry is trying to resolve copyright tension.
Automattic's board ousts CEO Matt Mullenweg
Matt Mullenweg, the co-creator of WordPress and CEO of Automattic (the company behind WordPress.com), was placed on a paid leave of absence by the company's board. Mullenweg said in an internal message that the move was against his will and that the CFO had worked with board members to force the decision. TechCrunch has the background on the forced leave, and The Verge covers Mullenweg's own account of what happened. Automattic's products underpin a significant portion of the web, so any leadership instability is worth watching for developers who build on WordPress.
LG denies its TVs eavesdrop for ads
Researchers published findings claiming they were able to exploit microphone and content-recognition features in an LG smart TV to capture nearby conversations. LG rejected the claim that its televisions record ambient audio for ad targeting. CNET has LG's denial and what the researchers actually found. The episode fits a broader pattern of consumer hardware being scrutinized for unexpected data collection, and it is a good reminder for security-focused learners that attack surfaces now include everyday home devices.
Four hacker groups shared the same Chrome and Windows exploit
Security researchers found that four separate threat groups were using identical exploit code targeting vulnerabilities in Chrome and Windows. The overlap suggests either coordination or a shared source for the toolkit. Researchers also noted that AI is accelerating the pace at which new vulnerabilities are discovered and weaponized. Ars Technica has the technical detail on the shared exploit kit. If you are studying cybersecurity, this illustrates how a single unpatched gap can be exploited by multiple actors at once, and why patch cadence matters so much.
Massachusetts adds clean-power rules for data centers
Massachusetts has joined a small but growing list of US states requiring new data centers to meet clean energy standards. The rules add constraints on where and how large AI infrastructure can be built. TechCrunch covers the Massachusetts regulations and the state-level trend. As AI demand drives data center construction to record levels, the regulatory environment around energy use is becoming a real factor in how and where tech companies scale their infrastructure.
The week's news covers a lot of ground, from the hardware choices on a folding phone to the policy decisions shaping AI infrastructure. Each of these threads will keep developing, and the AI-in-everything story is far from settled, whether that means math proofs, music models, or the watch on your wrist.
