Snap is introducing “Specs Intelligence,” a new AI assistant that can connect other digital accounts to help you with things like work tasks and keeping track of travel information. It seems similar to AI assistants like Meta’s Muse and Gemini’s Spark, though Snap is pitching Specs Intelligence as an “anticipatory AI service” that “helps you manage what needs attention today, so you can make progress toward your longer-term goals.” Like with other AI assistants, you can also chat with it.

Specs Intelligence is being launched alongside Specs, Snap’s first consumer pair of augmented reality glasses. But it’s also available on iOS today in preview, and you can get on a waitlist for the “full early-access experience” that’s coming to Mac. Snap wouldn’t say on the record how the preview differs from the full experience.

Here’s how Specs Intelligence works, according to Snap:

SPECS Intelligence builds an understanding of your goals, priorities, relationships, and routines from the apps and tools you choose to connect. The service is designed to carry that personal context across devices and deepens its understanding over time, helping you move between work, personal responsibilities, and the things you care about.

SPECS Intelligence is designed to anticipate when help would be useful and surface relevant information and actions without waiting for a prompt. Before a meeting, it could highlight decisions to make, questions to ask, and things to remember. For an upcoming trip, it could bring together flight, hotel, and dinner plans while flagging a work deadline that falls during the visit.

If you’re using Specs Intelligence on the Specs AR glasses, Specs can pull personal content and place it “directly into view” when relevant, like when you ask for a daily briefing before work. According to a video, Specs Intelligence can connect to apps like Gmail and Slack.

Specs Intelligence uses “a proprietary combination of open-source models hosted in the US alongside local LLMs,” according to spokesperson Cassie Bumgarner, who adds that “We regularly update the mixture of models as we continue iterating towards the best version of SPECS Intelligence.” But the company faces real challenges in that it needs users to trust it with their information and with data from connected services. Snap is trying to get ahead of those concerns by saying that it can’t view or access personal content you use with Specs Intelligence and that “Personal content from connected accounts will not be used to train or fine-tune Snap’s AI models, or to serve personalized ads.” But it’s still a tall order, especially given how we’ve found other AI assistants with broad access to data to be pretty unnerving.

Nintendo has gotten quite good at making open-ended games. The Legend of Zelda: Breath of the Wild upended the idea of what a Zelda game could be by giving the player free rein to tackle any part of the map, a philosophy iterated upon with Tears of the Kingdom. On the Switch 2 this idea expanded to Mario Kart World and Donkey Kong Bananza. Now, Nintendo has brought this philosophy to Fire Emblem with Fortune’s Weave. It works really well; with its many ways to progress through the game and tailor teams to your liking, Fortune’s Weave has become my favorite game on the Switch 2 so far.

Fortune’s Weave’s structure is a big reason why it feels so open: Instead of locking you to one path, Fortune’s Weave is designed so that you can experience multiple stories in one massive playthrough. There are four main hero routes and a final act that ties things together, giving you several opportunities to start over from scratch and master the dizzying amount of mechanics and meta-strategy involved in winning fights and building up your group. From early on you can swap between heroes and storylines at will, and it’s even possible to merge stats from characters you use in multiple routes to make one supercharacter later. The systems go really deep, but Fortune’s Weave gives you many ways to customize things exactly to your liking. 

The game is split into a few sections. There’s a brief prologue that introduces the main protagonist and your mission to save the world and defeat an evil demon lord. To aid in your quest, you can recruit the four missing heroes from the recent past by playing through their individual stories in two parts. Part one features the hero of your choice as they compete in a grand tournament in the city of Dagsion, while part two carries over your hero’s squad to fight in a war. If you complete a hero’s path, you can take them and any of their allies into part three, where the true protagonist leads an effort to fight the undead.

Parts one and three give you a lot of leeway to train up for battles and build your teams as you’d like. You’ll quickly fall into a rhythm of splitting time between the main story battles, running around Dagsion to prepare for fights, enlist allies, and explore a vast map to gather resources and fight in small skirmishes. There’s a calendar system that counts down the days toward the next main story event, but you have the freedom to spend your days however you’d like. I fell into a comforting pattern where I’d explore a map for one in-game week, come back to Dagsion to complete weekly activities like training up skills and sharing meals with people in town to build rapport, and then back head out to the map to tackle more quests or grind out some XP. (Part two is pretty much all about battles, as you’re mostly just playing one big fight after another with very little pre-battle prep.) 

The main battles are your typical Fire Emblem turn-based strategy. You can choose which allies you want to use, and then you spend each turn deciding how to best attack your enemies or buff your team with healing spells and other support moves. Each of your characters has a different class, and a diverse squad is key; as much as I’d love to run a team that consists just of flying units, if they came upon any archers, they’d probably be toast. Some battles have different objectives to mix things up, like tasking you with defeating thieves before they escape while simultaneously fighting giant birds. On the game’s standard difficulty, and with permadeath turned off, I didn’t have much trouble. But I often invoked the blessings of gods during battles, which give you handy abilities like rewinding back to a previous move if you make a mistake. 

There are also battles that play out like a more typical turn-based RPG, which you’ll run into in dungeons on the map and with occasional story battles. I found these much less interesting than the strategy battles, and thanks to an autoplay feature, I usually skipped them altogether to get the experience and stat boosts.

I chose to finish all four hero routes in parts one and two so that I could bring them into part three. It was a herculean effort: Completing each hero’s path took me about 100 hours in total. And that was after reducing battle animations, skipping cutscenes I had already seen, and doing a lot of fast traveling. But with each hero’s playthrough, I better understood how to train characters, assign them advanced classes to build balanced teams, and use resources from the map to make my weapons more powerful.

Part three also adds one extra complication: Instead of roaming the map with a single character and their team, you can juggle multiple squads at a time. This initially felt daunting, but I quickly settled into deploying my squads to different corners of the map to finish quests, leveling up my whole team, and crushing a lot of bad guys. 

Not every hour of my playthrough, which ended at about 130 hours, was well spent. In parts one and two, I repeated a lot of quests, cutscenes, and exploration. Part two features the same levels, just with the characters swapped around depending on which hero you’re playing as. In part three, managing multiple squads simultaneously can feel like you’re inching across the map. And the strange, leery tea party minigame from Three Houses returns as “bird time,” where you talk with and “observe” other characters as the game’s main protagonist — who has taken the form of a bird — to build support. I promise the bird form makes sense in the context of the story, but it’s still weird.

But even with the repetition, I found myself completely immersed in Fortune’s Weave. The loop of leveling up my characters, preparing for battle, and routing my enemies was nearly perfect, and I couldn’t get enough. Reaching part three in particular felt like a significant reward for my hard work; after 100 hours, I was thrilled when I could bring all of my teams together for the final push.

Fortune’s Weave is just the latest example of how Nintendo adapts the idea of openness in a way that makes sense for each franchise. In this case, it doesn’t change the core game like how Breath of the Wild transformed Zelda. Instead, the game’s open-ended nature adds even more layers of strategy to a series that had already started pushing its own boundaries with titles like Three Houses. Fortune’s Weave feels like the culmination of what Nintendo has been building Fire Emblem to be. 

^{Fire Emblem: Fortune’s Weave launches on the Nintendo Switch 2 on September 17th.}

Apple released macOS 27 Golden Gate on Monday, which brings the new Siri AI assistant, Liquid Glass improvements, improved performance, and more. The update is compatible with all Macs with Apple Silicon, but not Intel Macs — Apple released its final big update for Intel-based Macs last year with macOS Tahoe.

Siri AI is arguably the biggest feature in macOS 27, and it allows you to ask questions, search through things involving your personal context like photos and emails, and take actions in apps such as Messages and Reminders. There’s a new dedicated Siri app that keeps an archive of your chats for easier reference. However, Siri AI will initially launch as a beta and will only be available in English.

As for Liquid Glass, Apple says the update offers “uniform refraction and improved contrast,” and there is a new slider that makes Liquid Glass elements more or less opaque. And there’s one other major UI change worth celebrating: consistent window corner radii.