- AI Fire
- Posts
- 😱 ChatGPT Sleeps = We Panic
Read time: 5 minutes
OpenAI just released a brand new model (pro ver), that thinks slower? but smarter than anything else out there. It quietly dominated in math and science, leaving Claude 4 and Gemini 2.5 Pro in the dust.
What are on FIRE 🔥
IN PARTNERSHIP WITH BELAY
Economic pressure is rising, and doing more with less has become the new reality. But surviving a downturn isn’t about stretching yourself thinner; it’s about protecting what matters most.
BELAY matches leaders with fractional, cost-effective support — exceptional Executive Assistants, Accounting Professionals, and Marketing Assistants — tailored to your unique needs. When you're buried in low-level tasks, you lose the focus, energy, and strategy it takes to lead through challenging times.
BELAY helps you stay ready for whatever comes next.
AI INSIGHTS
OpenAI’s new o3-pro model just beat both Claude 4 Opus and Gemini 2.5 Pro in science and math tests. And yes, it's now the most powerful AI in their lineup. And it’s built for thinking, not just chatting. If you care about reliability over speed, this might be the model you switch to permanently.
🔍 What is o3-pro?
o3-pro is a new reasoning-focused AI model, upgraded from OpenAI’s o3.
It’s designed to think step by step, making it stronger in areas like: Coding + Math + Science + Academic writing + Business analysis
Replaces o1-pro in ChatGPT Pro and Team plans starting June 10.
Also available via API, priced at $20/M input tokens, $80/M output tokens.
🧠 What Makes It Different
Reasoning First → It was built to solve problems methodically, not just autocomplete text.
Review Scores → It outperforms both o1-pro and o3 across every tested area. Expert reviewers ranked it higher.
Benchmark Results:
AIME 2024 (math): Beats Gemini 2.5 Pro
GPQA Diamond (PhD-level science): Beats Claude 4 Opus
o3-pro has full tool access including: Browse the web + Use Python + Analyze documents + Process visual inputs + Leverage memory for personalization → But: responses are slower than o1-pro, due to deeper reasoning process.
Why It Matters: o3-pro is OpenAI’s answer to a growing problem: speedy AI models that hallucinate too much under pressure. This one is slower but sharper, and that’s exactly what researchers, and serious users have been asking for. OpenAI is splitting its lineup: GPT-4o for speed + multimedia, o3-pro for deep logic and trust.
PRESENTED BY AVIRON
Meet the Aviron Victory Treadmill—designed to keep you consistent and your workouts from collecting dust. With science-backed programming inspired by game design and industry-leading hardware that’s comfortable, powerful, and built to last, the Victory will help you stay on track and actually see results. Here’s what makes the Victory different:
Science-Backed Workouts: Designed using game psychology to keep you motivated, engaged, and coming back for more day after day.
Endless variety for every fitness goal: Play, train, stream, and compete with the most extensive workout library, all on a stunning 22" HD touchscreen.
A low to the ground, comfortable design: Walk, run, and train with Aviron’s cushioned ultra-low 4” Cloud Stride deck for advanced shock absorption and comfort. Commercial-quality build and performance: A 3.5 CHP motor, maintenance-free pre-waxed belt, 10-year warranty, and 400 lb capacity deliver unmatched durability and reliability. Ready to try a workout you won’t want to quit?
TODAY IN AI
AI HIGHLIGHTS
🔌 ChatGPT suddenly went down worldwide. Nearly 2,000 outage reports were logged on Downdetector, leaving thousands of users frustrated with error messages. This hiccup shows our heavy reliance on AI.
How much did the ChatGPT outage affect your work?Some teams paused entirely. Others barely noticed. |
🚄 Google just unveiled Veo 3 Fast, which can generate videos 2x faster while maintaining a 720p resolution and a bunch of serving optimizations. Try it here.
🥅 X users are dragging Apple for pointing out flaws in reasoning models in its latest report while still failing to launch its own LLM. The study did spark some interesting discussions.
📹 One Instagram account reportedly pulled 8M views in a day using Veo 3 for stormtrooper-style vlogs. Just turn everyone into a stormtrooper. Watch it here.
🤰 Doctors at Columbia University used AI to help a couple with 19 years of infertility finally get pregnant. It’s the first known case of pregnancy made possible through AI.
☁️ OpenAI has quietly signed a cloud deal with Google, though they’re direct AI competitors. This is an "arms dealer" play. Microsoft’s tight grip on OpenAI may be loosening.
💰 AI Daily Fundraising: Plug and Play has secured a $50 million Fintech & AI Fund with contributions from 9 institutional investors, including EXL. This fund will focus on AI financial services.
AI SOURCES FROM AI FIRE
NEW EMPOWERED AI TOOLS
🧠 Chronicle goes from raw thoughts to polished presentations easily
🛠️ Bubble builds and launches web and mobile apps without writing code
📄 Huntr is an AI powered line-by-line review of your resume in under 5 mins
🎞️ Blockies Animate has plug-and-play UI animations for modern web app
🌐 Covolute is an AI web creator on an infinite, real-time collaborative canvas
AI QUICK HITS
🍎 Here’s everything (AI) that Apple announced at WWDC 2025. Visual AI?
💵 OpenAI hit $10B in annual recurring revenue, roughly ~$833M a month
🛑 OpenAI’s first open model in years is delayed until later this summer
🌟 2025 CNBC Disruptor 50: See the list of companies leading new AI era
🧠 Microsoft-backed Mistral launched its first reasoning model to rival OpenAI
AI CHART
The UK government is partnering with Google to use Gemini AI to fix one of the most broken parts of the public sector: infrastructure planning. A new AI tool called Extract can scan messy, handwritten, or scanned planning documents and process them in seconds — a job that normally takes hours.
→ It’s now being tested with a nationwide rollout expected in 2026.
Extract converts physical and unstructured planning documents, even blurry maps and handwritten notes, into searchable, structured digital data.
It cuts 2 hours of human work down to 40 seconds, according to early trials. It’s designed to accelerate slow infrastructure and housing decisions in the UK.
The aim is to remove tedious manual work so planners can focus on actual decision-making.
Also helps reduce massive backlogs that have plagued housing and public works projects for years.
This is one of the most practical AI deployments yet in the public sector.
→ And we can see this move truly strengthens Google’s positioning in the enterprise and government AI space. Google directly supports the UK’s 1.5 million home-building target by unblocking slow processes.
AI JOBS
We read your emails, comments, and poll replies daily
How would you rate today’s newsletter?Your feedback helps us create the best newsletter possible |
Hit reply and say Hello – we'd love to hear from you!
Like what you're reading? Forward it to friends, and they can sign up here.
Cheers,
The AI Fire Team
Reply