In partnership with

Welcome to today’s AI Boost

A lean in-house chatbot shows Apple is finally moving beyond feature demos, while Chinese researchers unveil a neuromorphic machine that packs macaque-level neurons. Anthropic has stopped letting OpenAI probe Claude, Google is rolling out its Deep Think reasoning mode to Gemini subscribers, OpenAI lays out its vision for do-anything agents, and fresh numbers put the company’s revenue run-rate at twelve billion dollars. Details follow.

Learn from this investor’s $100m mistake

In 2010, a Grammy-winning artist passed on investing $200K in an emerging real estate disruptor. That stake could be worth $100+ million today.

One year later, another real estate disruptor, Zillow, went public. This time, everyday investors had regrets, missing pre-IPO gains.

Now, a new real estate innovator, Pacaso – founded by a former Zillow exec – is disrupting a $1.3T market. And unlike the others, you can invest in Pacaso as a private company.

Pacaso’s co-ownership model has generated $1B+ in luxury home sales and service fees, earned $110M+ in gross profits to date, and received backing from the same VCs behind Uber, Venmo, and eBay. They even reserved the Nasdaq ticker PCSO.

Paid advertisement for Pacaso’s Regulation A offering. Read the offering circular at invest.pacaso.com. Reserving a ticker symbol is not a guarantee that the company will go public. Listing on the NASDAQ is subject to approvals.

1. Apple’s stripped-down chatbot project surfaces

A small “Answers, Knowledge and Information” team inside Apple is building a lightweight conversational model that can run mostly on device, giving Siri a ChatGPT-style upgrade without heavy cloud costs. The prototype, nicknamed Apple Answer Engine, skips creative writing and focuses on concise responses drawn from on-screen context. Engineers hope to ship a preview alongside iOS 26 next spring, but the schedule could move if accuracy targets are missed.

2. “Darwin Monkey” packs two billion neurons into a brain-like supercomputer

Zhejiang University researchers revealed Darwin Monkey, a neuromorphic cluster built from 960 Darwin 3 chips that together simulate more than two billion neurons and one-hundred-billion synapses. The system already completes content generation and logic tasks using DeepSeek’s brain-inspired model and is hailed as the first computer with neuron counts close to a macaque. Developers say it will push research on brain-style AI while using just a tenth of the energy required by conventional GPU racks.

3. Anthropic cuts off OpenAI’s access to Claude models

Anthropic has revoked API keys that OpenAI engineers were using to benchmark Claude against GPT-5 in areas like coding and safety. Anthropic claims such competitive testing breaks its terms, though it will still allow limited calls for public leaderboard comparisons. OpenAI calls the cutoff disappointing but says its own o-series models remain available to Anthropic, highlighting growing rivalry as both chase enterprise customers.

Meet your new assistant (who happens to be AI).

Meet Skej — your new scheduling assistant. Whether it’s a coffee intro, a client check-in, or a last-minute reschedule, Skej is on it. Just CC Skej on your emails, and it takes care of everything: checking calendars, suggesting times, and sending out invites.

4. Google releases Gemini 2.5 Deep Think for AI Ultra subscribers

Deep Think, an enhanced reasoning setting inside Gemini 2.5, is now live for Google AI Ultra users and select academics. The mode extends “thinking time” so the model can explore parallel solution paths, delivering Bronze-level International Math Olympiad scores in minutes instead of hours. Google explains that new reinforcement-learning techniques help Deep Think choose among hypotheses and excel at code, science, and creative design tasks, all while posting safer outputs than 2.5 Pro.

5. Inside OpenAI’s push toward full AI agents

A TechCrunch deep dive traces OpenAI’s journey from MathGen research to the o-series reasoning models that underpin its ChatGPT Agent vision. Internal teams now focus on reinforcement learning methods that let multiple sub-agents search for answers before voting on the best result, a strategy that earned a gold medal at the 2025 International Math Olympiad. CEO Sam Altman says the goal is a computer that can complete any online task once asked, but the company still wrestles with hallucinations and complex workflows.

6. OpenAI annualized revenue hits twelve billion dollars

OpenAI has doubled revenue since January, pulling in about one billion dollars per month and counting roughly seven-hundred million weekly active ChatGPT users. The Microsoft-backed firm now expects to burn eight billion dollars in cash this year as it expands model training and builds massive data centers, and it is finalizing the second tranche of a thirty-billion-dollar funding round with Sequoia and Tiger Global among the key investors.

Turn AI from Party Trick to Profit Machine

I built Infinity for one goal: hand you the same AI systems and automations I lean on every day.

Plug them in, cut the busy work, and watch new revenue pop up.

You’ll quickly become the go to AI expert everyone trusts.

Ready to plug in?

How would you rate today's newsletter?

Vote below to help us improve the newsletter for you.

Login or Subscribe to participate

Stay tuned for more updates, and have a fantastic day!

Cheers,
Zephyr

Keep Reading

No posts found