You don't have to read everything — just the right thing. 1440's daily newsletter distills the day's biggest stories from 100+ sources into one quick, 5-minute read. It's the fastest way to stay sharp, sound informed, and actually understand what's happening in the world. Join 4.5 million readers who start their day the smart way.

Join for free today!

For the past two years, when people talked about the AI race, they meant OpenAI, Google, and Anthropic. Meta was the company that kept spending billions, kept promising breakthroughs, and kept showing up late.

That changed this week.

On April 8th, Meta's Superintelligence Group — the team CEO Mark Zuckerberg built after a multibillion-dollar overhaul that included recruiting Scale AI founder Alexandr Wang — launched its first public model: Muse Spark.

The reviews are mixed but the signal is clear: Meta is no longer watching from the sidelines.

What Is Muse Spark, and Why Does It Matter?

Muse Spark is Meta's first large language model developed entirely by its new in-house superintelligence team. It's being rolled into Meta's existing platforms — Facebook, Instagram, WhatsApp, and Meta's business tools — rather than being offered as a standalone product like ChatGPT.

That distinction matters. This isn't a product you sign up for. It's infrastructure baked into systems that roughly 3.5 billion people already use every month — including millions of NYC small business owners who run ads on Instagram and Facebook.

Initial benchmarks show Muse Spark is competitive with the current generation of frontier models in most categories. ArsTechnica called it "strong but acknowledges gaps." Bloomberg noted the model is already being used internally across Meta's advertising, recommendation, and content systems.

This is not a science project. Meta is deploying this in production, at scale, immediately.

The Money Meta Is Betting

To understand how serious Meta is, follow the cash.

This week, Meta and CoreWeave expanded their AI computing deal to $21 billion — up from the previous $14.2 billion agreement. That's money committed through 2032. Meta's CEO has publicly stated that underinvesting in AI infrastructure would be a mistake that couldn't be fixed later.

For comparison: a year ago, people were questioning whether Meta could compete with OpenAI at all. Today Meta is committing more infrastructure dollars than most entire countries spend on technology.

What This Means for NYC Small Business Owners

If you run a business and you advertise on Meta platforms — Facebook, Instagram, WhatsApp — here's what matters about Muse Spark:

1. Ad targeting is about to get sharper.
Meta's AI already runs the most sophisticated ad targeting engine in the world. Muse Spark is being integrated into that engine. That means better audience matching, more accurate interest targeting, and potentially lower cost-per-click for campaigns that are already working. But it also means the floor on average performance rises — you'll need sharper creative to stay ahead.

2. Content tools inside Meta are getting more capable.
Meta has been rolling out AI content generation tools for business accounts. With Muse Spark integrated, expect those tools — caption suggestions, image generation, ad copy drafts — to improve meaningfully. For small businesses that don't have a marketing team, this matters.

3. The competition for AI-assisted ads is intensifying.
Google has been running AI-enhanced campaigns (Performance Max) for two years. OpenAI is building commerce integrations. Now Meta has its own model improving its own ad platform. If you're not using AI tools to optimize your ads, your competitors who are will have a structural advantage.

4. Customer service automation on WhatsApp is expanding.
For NYC businesses that use WhatsApp for customer communication — which is increasingly common in immigrant-owned businesses in Queens, Brooklyn, and the Bronx — Meta's AI integration will enable more sophisticated automated response tools. This is coming regardless of whether you opt in; your competitors will be using it.

The Bigger Picture: Three-Way War, One Winner (For Now)

The current state of the AI model race looks like this:

Each of these companies is fighting for a different thing. Google wants to keep search and cloud. OpenAI wants to be the enterprise standard. Meta wants to own the social layer of AI — which means AI that's embedded in communication, advertising, and commerce.

For NYC small business owners, the practical translation is this: the platforms you already use are getting smarter, faster. The businesses that learn to work with these tools will see meaningful advantages within 12 months. The ones that wait will find the gap harder to close.

What Anthropic's New "Mythos" Model Adds to the Picture

It wasn't just Meta this week. Anthropic quietly gave Apple and Amazon early access to a new, more powerful unreleased model called Mythos — a model so capable that Anthropic is limiting its distribution specifically because of concerns it could be weaponized for cyberattacks.

Read that again: an AI company is restricting its own model release because it found security vulnerabilities in every major operating system and web browser during internal testing. Project Glasswing — Anthropic's cybersecurity-focused AI — has now been identified as a material threat to infrastructure by the company that built it.

The practical implication for NYC business owners: if you store customer data, process payments, or operate any web-based system, your cybersecurity posture matters more this year than it did last year. AI is making attacks cheaper and faster to execute, not slower.

The Short Version

Meta launched a competitive AI model. It's already in the platforms millions of NYC businesses use daily. It's going to make those platforms smarter, more competitive, and more capable — both as tools you can use and as systems you have to stay ahead of.

The businesses that engage with this proactively — learning the new ad tools, testing the AI content features, shoring up their cybersecurity — will be better positioned than the ones that treat this as background noise.

This race isn't slowing down. It's accelerating.

The Metro Intel covers AI, tech, real estate, and local business news for New York City residents and small business owners. If someone forwarded you this, you can sign up at themetrointel.com.