- Everyday AI
- Posts
- Celebrating Everyday AI's 500th Episode: What we've learned about GenAI so far and where we're headed
Celebrating Everyday AI's 500th Episode: What we've learned about GenAI so far and where we're headed
Live at Google Next: Google drops dozens of new AI treats at its Cloud Next conference, Claude unveils $200/mo Claude Max plan, Microsoft dials back AI investments and more.
Everyday AI
April 09, 2025
š Subscribe Here | š£ Hire Us To Speak | š¤ Partner with Us | š¤ Grow with GenAI
Outsmart The Future
Sup yāall š
Welp.
That wasā¦.. emotional apparently?
Whoops.
Thanks for your support to get us to 500 episodes. Yāall voted last week that you wanted our 500th episode of Everyday AI to be about our biggest AI takeaways since we started the show.
Soā¦. thatās what we did today.
AI t
Have any thoughts? Go share them. Iām going to get to all of your comments and questions soon.
Appreciate yāall!
āļø
PS ā I partnered with Google to bring yāall some AI scoop from industry leaders. Keep an eye out on this page on LinkedIn later tonight or YouTube once we announce our next show covering whatās new with Google Gemini.
Today in Everyday AI
9 minute read
š Daily Podcast Episode: 500 episodes in, this is how Generative AI has changed since Day 1. Give todayās show a watch/read/listen.
šµļøāāļø Fresh Finds: Walmart fighting tarrifs with AI, OpenAI's new Pioneers Program, NVIDIA dodges export restrictions and more.. Read on for Fresh Finds.
š Byte Sized Daily AI News: Live at Google Next: Google drops dozens of new AI treats at its Cloud Next conference, Claude unveils $200/mo Claude Max plan, Microsoft dials back AI investments and more. Read on for Byte Sized News.
š§ Leverage AI: Donāt have an hour to deep dive on todayās 500th show celebration? All good. Hereās the quick bullet points. Keep reading for that!
ā©ļø Donāt miss out: Did you miss our last newsletter? We talked about: Report: Elon's team using AI to spy on anti-Trump employees, Amazon makes big updates to its AI video model, White House wants more Chief AI Officers, Meta Benchmark drama and more.. Check it here!
The world is completely different than it was before ChatGPT š
GenAI has shook the world, and Everyday AI has been there to try and provide solid ground.
Every step of the way.
As we celebrate Everyday AI's 500th episode, we're taking a look back at how far the GenAI landscape has changed and talk about what's next.
Celebrating Everyday AI's 500th Episode: What we've learned about GenAI so far and where we're headed
Also on this pod:
Jordan gets a bit emotional š„¹
Why full-time employment wonāt mean much š¼
Big Techās hard pivot to AI everything š¦¾
Itāll be worth your 1 hour plus.
Listen on our site:
Click to listen
Subscribe and listen on your favorite podcast platform
Listen on:
Hereās our favorite AI finds from across the web:
New AI Tool Spotlight ā The Swarm uses AI to strengthen relationships, Helix is a collaborative AI coding agent, and the AI Command Bar turns your commands into actions.
Retail ā Walmart is fighting tariffs with AI and fashion trends.
AI Education ā Microsoft is promoting its free AI Skills Fest, where you can learn AI from the pros.
AI Startup ā OpenAIās new Pioneers Program helps startups fine-tune AI models and set industry benchmarks for real-world impactāwant in?
AI Developments ā Google is bringing The Wizard of Oz to the sphere with GenAI.
(Side note ā this was dope. We were there!)
announces at #GoogleCloudNext that @GoogleDeepMind is using GenAI to bring the Wizard of Oz the sphere
Giving the characters ānew lifeā and making history.
Thoughts?
ā Jordan Talks Everyday AI (@EverydayAI_)
12:20 AM ā¢ Apr 9, 2025
AI and the Economy ā Chinese social media has been mocking U.S. President Trumpās tariffs with an AI-generated video depicting Americans as sweatshop workers.
AI Video ā Googleās impressive Veo2 got upgraded with simple camera controls.
GPUs ā Nvidia's H20 chip dodges U.S. export restrictions after a Mar-a-Lago dinner, keeping China in the AI race. What happened?
Adobe and AI ā Adobe detailed its vision for agentic AI in creative spaces.
1. Google Unveils AI Powerhouse Chip Ironwood š ļø
In a major step forward for artificial intelligence hardware, Google introduced its seventh-generation AI chip, Ironwood, at a cloud conference on April 9, 2025.
Designed specifically for inference computingāthe backbone of applications like ChatGPTāthe chip doubles energy efficiency compared to its predecessor, Trillium. Capable of running in massive clusters of up to 9,216 chips, Ironwood is set to speed up AI applications while slashing operational costs. This bold move positions Google as one of the few serious challengers to Nvidia's dominance in AI processors
2. Anthropic Rolls Out Premium $200/month Claude Max plan š¤
Anthropic, the Amazon-backed AI startup, just unveiled āClaudeās Maxā plan, a new subscription tier starting at $100 per month, giving power users up to 20 times the usage compared to its $20 tier.
The move positions Anthropicās Claude chatbot as a direct competitor to OpenAIās ChatGPT Pro, which costs $200 monthly. Designed for professionals in coding, finance, and marketing, this tier also offers priority access to cutting-edge models and features like the soon-to-be-launched voice mode.
3. Google Unveils Gemini 2.5 Flash: AI Built for Speed and Efficiency ā”
Google has announced the launch of Gemini 2.5 Flash, a new AI model designed for cost-conscious, high-volume applications like customer service and real-time tools.
Prioritizing efficiency over cutting-edge accuracy, this āreasoningā model lets developers balance speed, accuracy, and cost for tailored performance. The move could shake up the AI market by offering a cheaper alternative to pricier flagship models while addressing demand for scalable solutions. Starting Q3, Gemini models will also be available in on-prem environments via Google Distributed Cloud, bolstered by Nvidiaās Blackwell systems.
4. YouTube Backs No Fakes Act to Fight AI Deepfakes š¤
YouTube has thrown its weight behind the No Fakes Act, a proposed federal bill aimed at banning unauthorized AI-generated clones of peopleās voices and images. According to Variety, Senators Chris Coons and Marsha Blackburn are leading the charge, with widespread support from SAG-AFTRA and major entertainment groups concerned about the rise of fake celebrity endorsements and AI impersonations.
The bill includes a takedown system, allowing platforms to swiftly remove flagged content, a feature YouTube is already testing with major creators like MrBeast. If passed, this legislation could set a new national standard for tackling deepfake abuse while leaving existing state laws intact.
5. Google Gemini 2.5 Pro Gets Smarter with Deep Research š§
Google is taking AI research to the next level with its Gemini 2.5 Pro (experimental) model, now featuring the advanced "Deep Research" capability.
The upgraded Deep Research tool scans hundreds of websites, builds tailored reports, and outshines OpenAI in metrics like comprehensiveness and writing quality. While currently limited to paid users, Googleās rapid rollout hints at broader access soon. For professionals needing detailed insights fast, this upgrade could be a game-changer in staying ahead of the curve.
6. Microsoft Hits Pause on AI Data Centers ļæ½*
Microsoft is tapping the brakes on several of its massive data center projects, including a $1 billion effort in Ohio, signaling a shift in its approach to AI infrastructure expansion.
According to reports, Microsoftās cloud operations president, the company is reevaluating its ambitious scaling strategy, citing the need for agility as demand for cloud and AI services evolves. Analysts suggest the slowdown may reflect adjustments in Microsoft's partnership with OpenAI, as the AI powerhouse pivots to build its own computing capacity.
Hundreds of guests.
Thousands of hours of work.
Todayās podcast is in-depth enough, so weāll try and make this recap quicker.
Hereās our biggest takeaways from 500 episodes ā and more than 2 years ā of keeping up with Generative AI together.
THE AI LANDSCAPE: TOTALLY UNRECOGNIZABLE š„
Two years ago it was just ChatGPT going viral. Today? EVERY enterprise has a gen AI strategy.
The technology that started as a little experiment has transformed into the driving force behind trillion-dollar companies, with giants like Bank of America throwing $4 billion into AI initiatives while traditional tech companies pivot their entire business models toward what they're calling "AI-first" approaches that are fundamentally reshaping how we think about computing.
THE FUTURE OF WORK IS AI-POWERED š©āš»
Working without AI will soon feel WEIRD. Like trying to do business without internet. Even non-technical people are starting to get it!
Your smartphone will probably run models more powerful than today's GPT-4 in just 2-3 years, completely transforming both our personal and professional lives as we shift from directly creating business value with our knowledge to orchestrating increasingly intelligent systems that can process, analyze, and generate content at speeds we could never match on our own.
BIG TECH'S ALL-IN PIVOT š¢
The six biggest companies by market cap are ALL from the same industry. First time EVER in US history! Microsoft, Google, Amazon, Nvidia, Meta - all AI companies now.
NVIDIA went from "that gaming card company" to sometimes-biggest-company-on-earth in just a few years because they're providing the computational foundation that makes all this possible, which shows exactly how massive this technological wave really is.
EMPLOYMENT SHAKE-UP AHEAD š¼
Let's not beat around the bush. AI will have a net negative impact on traditional full-time jobs. The gig economy that transformed ride-sharing is coming for knowledge work next.
Professional services will fragment into thousands of specialized micro-businesses charging a fraction of current rates, while large enterprises that successfully implement generative AI may see 50% productivity gains that unfortunately won't translate to four-day workweeks but rather to significant workforce reductions as companies optimize for profit rather than human wellbeing.
REDEFINING HUMAN SKILL & VALUE š¤
That thing you've been paid to do for 20 years? AI probably does it better. Sorry shorties. We're becoming tastemakers and curators rather than primary creators. It's weird when you first realize an AI is better at your job - that moment is coming for everyone if it hasn't already.
Instead of directly applying our expertise to solve problems, we're increasingly setting up the parameters for AI systems to do the heavy lifting while we provide the crucial judgment, context, and quality control that machines still struggle with despite their impressive computational abilities.
EXPERTISE NEEDS TO BE IN THE LOOP š
"Human in the loop" is just a comfort blanket. Sounds nice but fixes nothing. We need to think about "expertise in the loop" instead - actively directing and overseeing these powerful AI systems rather than passively monitoring them while sipping coffee.
Companies rushing to implement agentic AI need security experts, domain specialists, and ethical guardians actively participating in the process, because these systems are making increasingly consequential decisions that require not just any human oversight but specifically the right expertise at the right time.
THE AI RACE GOT COMPETITIVE š
Nine months ago OpenAI was the 90s Chicago Bulls - UNTOUCHABULL. Today? Google's Gemini 2.5 Pro is SHOCKINGLY good. Microsoft keeps dropping Copilot AI greatest hits weekly.
The competitive landscape has completely transformed from a one-player game to a heated race with multiple serious contenders, which means organizations need dedicated evaluation teams running weekly tests across all major models to determine which ones actually perform best for their specific needs rather than relying on general benchmarks that might not reflect real-world performance in their domain.
We all win in the end. But we lose sleep. lolz.
WHERE DO WE GO FROM HERE? š®
Agentic AI is taking over. Systems that work autonomously without human prompting. Slow, expensive sectors like consulting, legal, finance, and accounting are about to get ROCKED. Half will cling to old business models and lose clients from both ends of the spectrum - the top 20% who build internal capabilities and the bottom 20% who can't justify the expense anymore.
The other half will embrace AI and slash prices dramatically, while laid-off experts launch hyper-specialized boutique services charging 10% of legacy rates but delivering better results in a fraction of the time by leveraging the same AI tools that displaced them in the first place.
A PERSONAL THANK YOU š
This journey to 500 episodes has been WILD.
Not gonna lie - it's been exhausting. But every time someone says "that episode helped my company make a decision" or "you gave me the confidence to try AI" - it makes all those 5AM wake-ups worth it! (Orā¦ 4 AM like today.)
Massive thanks to everyone who's stuck with us during these sometimes-rambling shows, especially on days when we thought we bombed it. Special shoutout to my wife, whoās a friggin SAINT, and pushes me to help you all understand AI.
And thank you to YOU.
You make it all worth it. Every comment, email and DM (that I suck at responding to) is appreciated. Thatās been the fuel needed to keep this train choo-chooing.
So whaddya say?
Hereās to 500 more episodes? š„
Reply