Welcome to today’s AI Boost!
Growth, controversy and fresh hardware headline today’s lineup. ChatGPT is on the cusp of another usage milestone while Cloudflare accuses Perplexity of sneaking past robots.txt. Broadcom unveils a massive networking chip aimed at AI clusters, Google pokes fun at Apple’s delayed “Intelligence” launch, Nvidia scrambles to patch serious Triton flaws, and Google’s own AI bug hunter bags its first 20 security bugs. Read on for the full stories.
Better inputs. Sharper outputs. Download the guide to premium AI.
Building or refining generative AI models? This guide shows why scraped data falls short—and what to use instead. Learn how real-world behavior signals, clustering, semantic scoring, and visual diversity improve output. Plus, see how Shutterstock’s licensed data and services reduce risk and boost performance. Train smarter, faster, and more responsibly.
1. ChatGPT closes in on 700 million weekly users
OpenAI told TechCrunch the chatbot should cross 700 million weekly active users this week, up from 500 million in March and quadruple last year’s count. VP Nick Turley credited a March image-generation upgrade and said business subscriptions have reached five million seats. Sensor Tower data shows users open the app on average 12 days each month and spend 16 minutes per day, trailing only Google Search and X for engagement.
2. Cloudflare flags Perplexity’s stealth crawling tactics
A Cloudflare investigation claims Perplexity AI is rotating user-agents, IPs and even entire ASNs to reach pages that have explicitly blocked its declared bots. Tests on freshly created, no-crawl domains still produced detailed Perplexity answers, leading Cloudflare to de-list the startup as a verified crawler and add new blocking heuristics for its customers.
3. Broadcom’s new Jericho4 chip targets multi-site AI clusters
Broadcom’s Jericho4 switch silicon, built on TSMC’s 3-nm process, links data centers up to 60 miles apart and can scale to about 4,500 chips per system. The part uses high-bandwidth memory to buffer traffic bursts and encrypts data in transit, features aimed at hyperscalers wiring thousands of Nvidia or AMD GPUs together for training runs.
4. Google ad mocks Apple’s delayed “Apple Intelligence” rollout
In a 30-second Pixel 10 teaser, Google advises buyers to “switch” if they purchased an iPhone 16 for the promised Apple Intelligence upgrade that never arrived. The Verge notes Apple pulled its own ad and now cites Siri architecture issues for the slip, while Google’s spot lands ahead of the Pixel 10 launch on August 20 with many specs already leaked.
5. Critical bugs found in Nvidia Triton Inference Server
Security researchers at Wiz disclosed three CVEs in Triton’s Python backend that can be chained for remote code execution without authentication. Nvidia’s August bulletin patches these plus three other critical flaws, warning that unpatched servers risk model theft, data leaks and full compromise. No in-the-wild exploits are known yet, but admins are urged to update immediately.
6. Google’s AI bug hunter “Big Sleep” uncovers 20 flaws
Google says its DeepMind-built Big Sleep agent, reviewed by Project Zero staff, has reported 20 previously unknown vulnerabilities in popular open-source projects such as FFmpeg and ImageMagick. Each bug was discovered and reproduced autonomously before a human validated the finding, marking what Google calls a “new frontier” in automated vulnerability research.
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?
Stay tuned for more updates, and have a fantastic day!
Cheers,
Zephyr