The overarching theme of today’s news cycle is the highly anticipated rollout of Apple’s public betas for iOS 27, iPadOS 27, and macOS Golden Gate, bringing the first wave of Siri AI to eager public testers. Meanwhile, corporate drama continues to unfold as OpenAI publicly pushes back against Apple’s recent accusations of trade secret theft regarding its upcoming AI hardware.
Riot Games has officially announced League of Legends Classic, rolling the clock back over 15 years to bring players the original Summoner’s Rift, a 60-champion starting roster, and the glorious return of old-school runes and masteries. Launching July 29, this heavily requested mode promises a nostalgic trip for MOBA veterans while blending in essential modern quality-of-life updates like improved matchmaking and server infrastructure.
Halo Studios Reportedly Cancels New Multiplayer Project · IGN Microsoft’s flagship sci-fi franchise is taking another hit, as a massive new Halo multiplayer project codenamed ‘Ekur’ has reportedly been scrapped. Developed by Certain Affinity, the title was planned as a large-scale battle experience utilizing Unreal Engine, but development troubles on Campaign Evolved forced a tragic reallocation of studio resources. It’s a bitter pill for competitive Spartans who are desperately craving a fresh PVP experience to tide them over until the next console generation.
What will be left for us to work on? Princeton professor Arvind Narayanan’s ICML keynote takes a pragmatic axe to the existential dread surrounding AI and software engineering. Instead of AGI instantly displacing the workforce, he argues we are seeing a shift where AI compresses the middle “execute” layer of coding, while expanding the “decide” (spec/planning) and “deliver” (maintenance/accountability) layers. The most poignant takeaway for the technical crowd: our roles are fundamentally shifting from building systems to evaluating them, much like the transition from rowing a boat to steering a modern ship.
New York has become the first US state to enact a moratorium on new hyperscale data centers, pausing construction to address the AI-driven strain on the power grid and environment. The landmark decision signals a coming collision between the tech industry’s insatiable demand for energy and the physical reality of local infrastructure.
It’s been a chaotic rollercoaster of a week for Naughty Dog fans, experiencing extreme highs and devastating lows. First, we got the heartbreaking confirmation from the game director that the highly anticipated, but ultimately cancelled, The Last of Us Online was nearly 80% complete before Sony pulled the plug to forcibly refocus resources on single-player titles. However, the studio quickly softened the blow for its community; Neil Druckmann teased that there are still stops left on the road ahead for The Last of Us Part 3, while Naughty Dog’s creative director fueled intense speculation that Uncharted 5 is in development by posting research trip photos from a fort in Trinidad.
This week’s news was dominated by concrete leaks surrounding the highly anticipated foldable “iPhone Ultra” and the massive market success of the budget-friendly MacBook Neo. On the software and AI fronts, Apple deployed critical fixes for Apple Intelligence and iCloud, while reportedly preparing a standalone, Gemini-powered Siri app for iOS 27.
Foldable “iPhone Ultra” Enters Trial Production · Hardware Leaks Apple’s highly anticipated foldable device, tentatively named the “iPhone Ultra,” has reportedly entered trial production and is slated for a September launch. Leaks reveal an ultra-thin 4.5mm titanium, passport-style chassis that sacrifices Face ID for a side-button Touch ID, utilizing exclusive Samsung OLED panels. The premium device is expected to command a price tag crossing the $2,000 threshold.
The week was dominated by a frantic escalation in the global AI computing arms race, contrasting the boundless ambitions of billion-dollar infrastructure projects with the harsh realities of hardware bottlenecks and ecosystem crackdowns. As geopolitical tensions surrounding semiconductor supply chains intensified, major US AI labs aggressively consolidated their platforms, while domestic Chinese tech firms capitalized on the shifting landscape to push “de-CUDA-ization” and secure critical homegrown hardware.
Meta’s launch of Muse Spark marks a massive strategic shift, as the newly formed Meta Superintelligence Labs abruptly abandons the company’s recent open-weights strategy. By releasing a proprietary, natively multimodal reasoning model equipped with “Contemplating mode,” Meta is signaling its intent to directly rival extreme test-time reasoning systems like Gemini Deep Think and GPT Pro.
Meta · Muse Spark Meta introduced Muse Spark, its first major model since Llama 4, built on a completely overhauled data pipeline, architecture, and infrastructure. Keeping the model proprietary is a massive pivot to compete in the high-end reasoning space, with the company deploying it exclusively via the Meta AI app and an upcoming private API.
Pokémon Champions is shaping up to be a textbook example of a botched launch. The highly anticipated free-to-play hub meant to replace Scarlet and Violet as the definitive competitive platform launched globally this week to immediate fan backlash. Players are furious over missing features like local wireless, a pitiful launch roster of only 185 Pokémon, and brutal mobile-style gacha mechanics that gatekeep crucial competitive items. To make matters worse, a bizarre bug is forcing Switch 2 players to physically undock and redock their consoles just to escape a halved 1080p resolution, leaving the title feeling like an unfinished beta rather than the future of the franchise.
[Why, and how you need to sandbox AI-Generated Code? — Harshil Agrawal, Cloudflare] from the AI Engineer channel is the single best watch this week because it strips away agent hype to deliver a stark reality check: executing generated code means running untrusted internet code in production. It provides a strict, capability-based security framework for deciding when to use V8 Isolates versus full Linux containers to prevent compute exhaustion and credential leaks.