- Everyday AI
- Posts
- GPT-5 Released: 7 big trends you should know
GPT-5 Released: 7 big trends you should know
Microsoft Copilot 3D now available, Google Finance gets AI update, OpenAIās o3 beats Grok at chess and more!
š Subscribe Here | š£ Hire Us To Speak | š¤ Partner with Us | š¤ Grow with GenAI
Outsmart The Future
Today in Everyday AI
6 minute read
š Daily Podcast Episode: OpenAI just released GPT-5āushering in game-changing AI capabilities overnight. Discover whatās new inside GPT-5, how free users win big, and 7 trends reshaping business and productivity. Give it a listen.
šµļøāāļø Fresh Finds: Meta acquires WaveForms, Browser Company launches Ai browser subscription and California advances AI efforts. Read on for Fresh Finds.
š Byte Sized Daily AI News: Microsoft Copilot 3D now available, Google Finance gets AI update and OpenAIās o3 beats Grok at chess. For that and more, read on for Byte Sized News.
š§ Learn & Leveraging AI: OpenAIās new o3 and o4 models are powerful. But does that make them the best? Hereās everything you need to know. Keep reading for that!
ā©ļø Donāt miss out: Did you miss our last newsletter? We talked about OpenAI launching GPT-5 to all users, Apple losing AI talent, Grokās AI video generator now free to U.S. users and more. Check it here!
GPT-5 Released: 7 big trends you should knowš”
Obvious: ChatGPT's GPT-5 is here and it's really good.
Not so obvious: the gap between novice and experts just shrunk 90%.
In a short few hours, OpenAI gave free users access to the now world's most powerful model.
As the most used AI chatbot in the world by a wide margin, the quality work we all produce has also just gotten a huge bump.
But there's a lot beneath the surface.
Join us as we dissect what's new in GPT-5 and 7 big trends you probably don't know but should pay attention to.
Also on the pod today:
⢠GPT-5 Benchmark and Accuracy Improvements šÆ
⢠GPT-5 Vibe Coding and Canvas Features š§āš»
⢠Microsoft Copilot Instant GPT-5 Upgrade š
Itāll be worth your 49 minutes:
Listen on our site:
Subscribe and listen on your favorite podcast platform
Listen on:
Hereās our favorite AI finds from across the web:
New AI Tool Spotlight ā Chance AI is AI-powered visual search, Tanka is an AI messenger with smart reply and Webdraw lets you explore and build AI apps with 50+ models.
Meta ā Metaās superintelligence AI team is now called TBD Lab.
Meta has acquired AI audio startup WaveForms
AI Browsers ā The Browser Companyās AI browser now has a $20 subscription.
AI in Governance ā California Governor Newson is collaborating with Big Tech companies like Google and Microsoft to boost Californiaās AI future.
1. Microsoft Opens Copilot 3D to Everyone š
Microsoft has rolled out Copilot 3D as a free tool that converts 2D images into GLB 3D files, usable in 3D viewers, design apps, and game engines ā available now for anyone to try. According to Tom Warren at The Verge, the feature is timely as interest in quick, accessible 3D content creation surges, but it struggles with animals and humans, producing inconsistent results.
For creators and small teams, that means a fast, no-cost path to prototype assets and learn 3D workflows, though human- or creature-focused projects will still need traditional modeling or more specialized tools.
2. Google supercharges Finance with AI, live charts and real-time feeds š§āš«ļø
Google rolled out an AI-powered update to Google Finance that lets users ask complex finance questions and receive comprehensive AI answers with source links, plus advanced charting (candlesticks, moving-average envelopes) and expanded market data including commodities and more crypto.
The update also adds a live news feed and real-time data, aiming to keep users on Google rather than sending them to third-party services or chatbots for deep financial queries.
3. OpenAI's o3 beats Musk's Grok in Kaggle AI chess final āļø
OpenAIās o3 model went unbeaten to defeat xAIās Grok 4 in the final of a Kaggle AI chess tournament, with Googleās Gemini taking third place. The contest tested everyday conversational AIs on chess, a classic benchmark for strategic reasoning, and exposed surprising blunders from Grok that cost it the title.
That matters because these modelsā relative strengths in reasoning and planning can signal which platforms may best assist professionalsāhelping with problem solving, coding, or decision supportāso firms and jobseekers should watch vendor progress closely.
4. Tesla Shutters Dojo Team as Leader Exits šŖšŗ
Tesla is disbanding its Dojo supercomputer team and its leader Peter Bannon is leaving, with remaining engineers reassigned and many moving to a new startup, DensityAI, Bloomberg reports. CEO Elon Musk told staff the company will rely more on partners like Nvidia, AMD and Samsung for chips and compute, shifting focus to in-vehicle AI chips āexcellent for inference and at least pretty good for training.ā
The move scraps a longātouted ināhouse advantage that analysts once valued as a potential marketācap driver, and signals Tesla is betting speed and partner ecosystems to beat bespoke hardware development right now.
5. OpenAI hands out big one-time awards ahead of GPT-5 launch āļø
CEO Sam Altman told staff a day before GPTā5ās rollout that OpenAI will give special oneātime awards to roughly 1,000 technical employees in applied engineering, scaling, and safety, with top researchers in the mid singleāmillion range and engineers averaging hundreds of thousands.
The payouts will be distributed quarterly over two years and can be taken in stock, cash, or both, while OpenAI also plans to let more employees sell vested shares at a higher price than the earlier $274 round valuation.
6. AWS Faces Heat Over AI Usage Caps on Anthropic Models š¤
AWS is under scrutiny after reports from The Information reveal customers hitting frustrating rate limits using Anthropicās AI models via AWS Bedrock, with some calling the caps āarbitrary.ā
While some users complain of frequent error messages, AWS insists these limits are designed to ensure fair access, not due to server shortages. This controversy arrives amid Amazonās massive $8 billion investment in Anthropic and its aggressive $26 billion AI spending plan for 2024-25.
š¦¾How You Can Leverage:
A huge chunk of the world just upped their capabilities overnight.
Thatās because OpenAI just released its highly anticipated GPT-5, which is now the worldās most powerful and capable LLM.
The real power up though?
Casual users are gonna straight up superhero their LLM outputs due to GPT-5ās auto routing that automagically decides when it should crank out compute juice.
700 million ChatGPT users powered up overnight, as GPT-5 is out and available to all.
Even freebie users.
So naturally, today on Everyday AI we gave GPT-5 the quick rundown, unveiling whatās new and 7 HUGE trends you should pay attention to.
Hereās the quick and skinny.
1 ā Free Users Just 10xād Your Competition š
The floor shot through the previous ceiling.
Weāre talking about casual users going from GPT-4O (which honestly sucked) to a hybrid reasoning model that thinks, plans, and strategizes. Thatās not a 20% bump. Thatās a 5-10x capability explosion.
Your intern just got PhD-level analysis capabilities. Your competitorsā bootstrap teams just became strategy powerhouses. Every freelancer, contractor, and scrappy startup now has reasoning abilities that enterprises paid millions to access.
The capability gap between big budgets and no budgets just evaporated. When 700 million people wake up with exponentially better AI, disruption doesnāt trickle down from Fortune 500 boardrooms.
It floods up from everywhere else.
Try This:
Audit your teamās current AI usage immediately. If theyāre still prompt-engineering their way through basic tasks or avoiding AI entirely, theyāre about to be outmaneuvered by competitors who embrace GPT-5ās reasoning.
Pick one department and mandate strategic GPT-5 adoption this week. The performance gap is about to become embarrassing.
2 ā Your Pocket PhD Just Got Advanced Voice š£
You can now upload every business document, strategy deck, and data file into a custom GPT, then literally talk to it like youāre brainstorming with your smartest advisor. Ask it about podcast download trends, revenue patterns, or strategic decisions - and it sounds exactly like talking to a human expert.
This isnāt Siri reading you search results. This is having a PhD-level strategist who knows your entire business intimately and never forgets a detail.
Every executive is about to have a personal AI strategist that knows more about their company than their C-suite does.
Try This:
Create a custom GPT with your key business documents uploaded. Enable Advanced Voice mode and spend 10 minutes talking through your biggest strategic challenge. The conversational intelligence will shock you.
Roll this out to your leadership team within two weeks - this is your new competitive weapon.
3 ā API Pricing Wars Just Destroyed Claude š°
7.5x more expensive for inferior performance.
GPT-5 API costs $10 per million tokens. Claude Opus 4.1 costs $75 for the same output. Itās not even close on benchmarks, and the human preference tests arenāt competitive either.
Every company building AI features just saw their development costs plummet while their capabilities skyrocketed. Startups can now afford enterprise-grade reasoning. Small teams can outspend big budgets through pure efficiency.
Meanwhile, companies locked into Anthropic contracts are paying premium prices for B-tier performance. The economics just shifted dramatically, and itās not subtle.
Try This:
If youāre building AI features or considering development, get GPT-5 pricing quotes immediately. Calculate total development costs against current vendors. The savings will likely justify accelerating every AI project on your roadmap while competitors debug overpriced alternatives.
4 ā Microsoft Can Win Enterprise AI š
100 million Copilot users upgraded without lifting a finger.
While you debated AI budgets, Microsoft embedded the worldās most advanced reasoning model directly into Word, Excel, Teams, and every Office application. Every email, spreadsheet, and presentation just became exponentially more intelligent.
Plot twist: Most enterprise customers didnāt even know they could access reasoning models in Copilot before this. They were using inferior GPT-4O while paying premium licenses.
Microsoftās enterprise moat just became a canyon. Your productivity stack invisibly upgraded itself while competitors are still evaluating standalone AI tools.
Try This:
If youāre paying for Microsoft 365 Copilot licenses, verify your IT team enabled GPT-5 access today. If youāre not on Copilot, calculate productivity differentials between your team and competitors who just got AI-powered everything. The operational gap is about to become insurmountable.
5 ā Every App You Use Just Got Smarter š§
Your software stack invisibly upgraded itself.
Banking apps, CRMs, customer service bots, data analysis tools - hundreds of applications using OpenAIās API are switching from GPT-4O to GPT-5 right now. Hallucination rates dropped 45%. Medical question accuracy jumped from 87% to 98.4%.
You didnāt pay extra. You didnāt change vendors. Your existing software investments just compounded automatically while competitors using Anthropic APIs pay 7.5x more for inferior results.
Every piece of software with AI capabilities just became dramatically more accurate, reliable, and useful. Your tech stack is about to make you look genius-level smart.
Try This:
Contact your key software vendors today asking when theyāre upgrading to GPT-5 APIs. For critical applications still on older models, this conversation accelerates your competitive advantage by months. Document which tools upgrade first - they become your strategic weapons.
6 ā Non-Technical Teams Are About to Build Solutions š
Marketing managers building custom analytics dashboards. Operations teams creating workflow automation tools. Finance developing specialized calculators that integrate with existing systems.
Weāre not talking basic automation. These are functional applications solving specific business problems without IT bottlenecks. When Jordan cloned Slack in 30 seconds this morning, that wasnāt showing off - thatās the new baseline capability.
Even Cursor, the vibe coding king, just made GPT-5 their default model. When the pros switch, you know the capability shift is real.
Try This:
Pick one repetitive manual process your team complains about weekly. Open GPT-5ās Canvas mode and type ābuild me a tool that [solves specific problem].ā
Spend 30 minutes iterating. Youāll have a working prototype eliminating hours of weekly busywork. Scale this across departments immediately.
7 ā Enterprise AI Holdouts Just Ran Out of Excuses š¬
The last barriers to full deployment evaporated overnight.
Hallucination concerns? Cut in half. Accuracy issues? Medical errors dropped from 13% to 1.6%. Reliability problems? Auto-routing handles complexity automatically. Integration challenges? Your existing software is already upgrading.
96% of executives call AI a strategic priority, but only 4% have scaled it company-wide. The technical limitations theyāve been hiding behind just disappeared.
Early adopters establish unassailable advantages while late movers debug yesterdayās problems. The window for ācautious AI strategiesā slammed shut.
Try This:
Schedule an emergency AI strategy review within 48 hours. Identify which pilot programs can immediately scale organization-wide using GPT-5ās improved reliability. Your competition is already planning full deployment. Hesitation isnāt strategic anymore - itās surrender.
Reply