Get tomorrow's brief in your inbox
June 5, 2026
Today: Apple is unveiling a Siri overhaul with Google Gemini integration at WWDC on Monday, while Poke became the first AI agent approved for iMessage. Supabase hit a $10.5B valuation riding the wave of AI-assisted coding, and Meta is building data centers in tents to keep pace with demand.
Apple is betting its AI strategy on Gemini, and you'll see the results on Monday. At WWDC 2026, Apple is expected to announce a major overhaul of Siri that transforms it from a barely functional voice assistant into something that actually competes with ChatGPT and Claude. The key ingredient? Google's Gemini technology powering Siri's contextual understanding, multi-step task handling, and natural conversation abilities.
This matters because Apple controls the devices most small business owners use every day. If Siri can finally handle tasks like "find the contract we discussed last week and email it to John" or "schedule a meeting with everyone who was on the last sales call," that's automation that just works without adding another tool to your stack. Apple is also rumored to be opening an AI agent app store, which means you'll soon be able to delegate tasks to specialized agents right from your iPhone.
For businesses, this is the moment AI assistants graduate from party trick to practical tool. If Apple delivers, you'll have genuinely useful AI built into the devices you're already carrying. Watch Monday's announcement at 1 p.m. ET on Apple's website.
Poke becomes the first AI agent on iMessage. Apple approved Poke, an AI assistant that works through text messages, to run on its Messages for Business platform. You can use it for calendar management, daily planning, smart home control, and photo editing, all via iMessage. It already works over SMS and Telegram (100 million messages relayed so far), but iMessage access is significant because it puts AI assistance directly into the messaging app most iPhone users already have open.
If you want an AI assistant that doesn't require learning a new app, try Poke. The catch? Apple charges Poke a per-user fee, which means pricing will eventually reflect that distribution cost. No free tier details are available yet.
Meta launched a Facebook creator assistant. Meta's new AI assistant for Facebook creators gives personalized recommendations on when to post, what's trending, and what your audience is saying. It can analyze your performance without forcing you to parse dashboards, and it suggests content ideas based on trending audio and cultural moments. Rolling out now in the U.S., Canada, and India.
If you manage a Facebook page for your business, this is worth testing. It keeps you inside Meta's ecosystem instead of switching to ChatGPT for brainstorming, and it could help you post more consistently. Over 500 million users are now watching AI-translated videos weekly on Facebook, so Meta is doubling down on keeping creators engaged.
Hello Robot's Stretch helps people with mobility challenges regain independence. Hello Robot released Stretch 4, a home assistance robot designed to work in real homes with real people. It has a telescoping arm, wheeled base, and voice control via iPhone. One user who became quadriplegic in 2021 now uses it to make protein shakes independently, a task that previously required help from another person.
This is not a humanoid robot promising to replace all labor. It's a practical tool for people with mobility challenges to regain autonomy at home. If you run a home care business or serve clients with accessibility needs, Stretch represents the kind of assistive technology that's actually deployable today.
AI coding tools are making databases grow faster than ever. Supabase raised $500 million at a $10.5 billion valuation, doubling its value since October. The company makes backend tools for building AI apps, and it credits the surge to "vibe coding," where people without technical skills can build apps through simple text prompts using tools like Claude Code and OpenAI's Codex. Claude Code is now the largest contributor to databases on Supabase's platform.
If you've been putting off building an internal tool because you don't have a developer on staff, this is the moment to try AI-assisted coding. Supabase provides the database, authentication, and infrastructure. Claude Code or Cursor can write the app. You handle the business logic and design.
AI is causing real operational failures at major brands. Starbucks retired an AI inventory tool after frequent mistakes, and Pizza Hut's AI delivery system allegedly lost a franchisee more than $100 million in sales. This is a reminder that deploying AI in operations requires rigorous testing before you go live. The cost of getting it wrong can be existential for franchisees and small businesses.
If you're considering AI for inventory, logistics, or customer-facing operations, start with a limited pilot. Test it in a low-risk environment where failures are visible and reversible. The brands getting burned are the ones who assumed the technology would just work at scale.
Apple's AI agent fee structure signals a new cost of distribution. Poke's approval on iMessage comes with a per-user fee that Apple charges the startup. The exact pricing isn't public, but it's reportedly lower than Meta AI's fees on WhatsApp (which increased in response to EU regulation). If Apple opens an AI agent app store as rumored, expect similar per-user tolls to become standard.
For small businesses, this means AI tools that integrate with Apple's ecosystem may carry hidden distribution costs that eventually show up in subscription pricing. Keep an eye on whether tools you rely on start charging more after gaining iMessage or App Store access.
AI leaders are asking Congress to regulate synthetic DNA screening. Anthropic's Dario Amodei, OpenAI's Sam Altman, Microsoft's Mustafa Suleyman, Meta's Alexandr Wang, and Google DeepMind's Demis Hassabis signed an open letter urging lawmakers to require companies that sell synthetic DNA and RNA to screen purchases for sequences that could be used to develop biological weapons. The fear is that AI tools could make it easier for people without specialized expertise to design harmful pathogens, order genetic material online, and assemble it in a lab.
This is a rare moment of agreement across rival companies. If you work in biotech or adjacent industries, mandatory DNA screening could become a compliance requirement in the next 12 to 18 months.
Airbnb's Brian Chesky is launching his own AI lab. Chesky plans to back a new AI lab focused on user interaction and design. He'll remain Airbnb's CEO and won't lead the lab himself. Chesky has been advising Sam Altman since OpenAI took off and helped broker Altman's return to power after his board fired him. Now he's entering competition with his mentee's company. The lab's focus suggests Chesky believes existing frontier models aren't solving interface and design problems well enough.
Mira Murati is back, carefully. OpenAI's former CTO, now CEO of her own company Thinking Machines Lab, gave her first major media appearance in 18 months. She previewed "interaction models" that process continuous streams of audio, text, and video in 200-millisecond intervals, designed to pick up on interruptions, mid-thought corrections, and pauses in something closer to real time. No release date yet.
Meta is building data centers in tents. Meta constructed six tent-based data centers outside New Albany, Ohio, to cut construction time in half. The structures house billions of dollars worth of AI chips. The approach borrows from Tesla's parking lot tents during Model 3 production and xAI's modular gas turbine strategy. Meta plans to spend up to $145 billion on data centers and capital expenditures, and Wall Street isn't happy (stock down 5% this year). Tents are one way to trim the bill.
TSMC says it can't keep up with AI chip demand. Taiwan Semiconductor Manufacturing Co., the world's biggest semiconductor maker, is struggling to meet demand from American customers despite its U.S. factory buildout. CEO C.C. Wei said it could take a "very long time" to fulfill customers' needs with U.S.-based production. TSMC has already opened a factory in Arizona and plans to invest $165 billion to build three additional plants, alongside two advanced packaging facilities and an R&D center. If you're planning infrastructure or hardware purchases that depend on AI chips, expect constrained supply and rising prices to continue for years.
Kevin O'Leary agreed to halve his Utah data center project. The Shark Tank star's Project Stratos will shrink from 40,000 acres to 20,000 acres after mounting pressure from residents and activists. Even at half the size, it will still cover an area larger than Manhattan. Utah Senate President J. Stuart Adams pushed for a 75% reduction and asked O'Leary to implement water-minimizing technology and divert excess water to the shrinking Great Salt Lake. Data centers this large pose major concerns around energy usage, environmental impact, and pollution.
Test AI coding if you've been putting off building internal tools. With Supabase handling the backend and Claude Code or Cursor writing the frontend, you can prototype apps for scheduling, inventory tracking, or customer management without hiring a developer. Start with a problem that costs you 5+ hours a week of manual work. Use AI to build a basic version, then refine it based on feedback.
Set up automation between LinkedIn ads and your CRM. If you're running LinkedIn ads, read Zapier's guide on optimizing from ad to deal. Most B2B teams set up LinkedIn's Conversions API once and never revisit it. Connecting conversion events to your CRM and sales pipeline can dramatically improve attribution and ad performance.
Three patterns are colliding this week: Apple is making AI assistants mainstream with Siri's Gemini-powered overhaul, infrastructure companies are racing to meet chip and data center demand they can't fully satisfy, and businesses are learning the hard way that AI deployments require more testing than they thought. The next six months will separate companies that implemented AI thoughtfully from those that deployed it fast and broke things. Watch Apple's WWDC keynote on Monday for signals on whether consumer AI finally delivers on its promises.