- Everyday AI
- Posts
- Why AI Thinking Beats Traditional Business Strategy
Why AI Thinking Beats Traditional Business Strategy
OpenAI's flex processing, Gemini 2.5 Pro safety issues, Meta's machine perception AI tools, adopting AI business strategy and more!
Everyday AI
April 18, 2025
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: An AI mindset is crucial for modern business strategy. Learn how AI thinking could transform your company's approach in an ever-evolving landscape. Give it a listen.
šµļøāāļø Fresh Finds: Google demos AI glasses, OpenAIās interest in Cursor and AWS releases new framework. Read on for Fresh Finds.
š Byte Sized Daily AI News: OpenAI cuts costs with flex processing, Gemini 2.5 Pro safety raises concerns and Metaās new AI tools for machine perception. For that and more, read on for Byte Sized News.
š§ Learn & Leveraging AI: Hereās how you can ditch traditional strategy and adopt an AI strategy for success. Keep reading for that!
ā©ļø Donāt miss out: Did you miss our last newsletter? We talked about OpenAI's safety reasoning monitor, Copilot Studio getting Computer Use and how the U.S. may ban DeepSeek. Check it here!
Why AI Thinking Beats Traditional Business Strategy š§ ļø
Youāre overthinking.
Meet the thing thatās outthinking you: AI.
While youāre stuck in strategy meetings and five-year plans, AI is making moves in seconds.
No ego. No guessing. Just results.
Old-school strategy? Itās slow. Itās flawed. Itās toast.
Want to stay relevant? You better learn how AI thinksāand fast.
Also on the pod today:
ā¢ Overcoming Inertia with AI Thinking ā¤“ļø
ā¢ AI Tools for Small and Large Businesses š ļø
ā¢ Evolution of Machine Learning Accessibility āļø
Itāll be worth your 27 minutes:
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 Omakase.ai turns your websites into Voice-powered sales agents, Universal Memory MCP makes your memory available to all LLMs and Shotup AI learns and remembers your screenshots.
Google ā Google has demoed its AI glasses at a recent TED talk.
Google One Premium is free for college students until Spring 2026, giving students free access to all of Googleās AI products. That means more than 20 million U.S. students now have access to Googleās premium AI offerings for free.
Wonder how many people are gonna take one class for that .edu email. Lolz
ā Jordan Talks Everyday AI (@EverydayAI_)
4:31 PM ā¢ Apr 18, 2025
OpenAI ā OpenAI had interested in buying Cursor before it ultimately acquired Windsurf for $3B.
Amazon ā Amazon has unveiled its AWS Well-Architected Generative AI Lens, a best practice framework for designing and operating GenAI workloads on AWS.
AI Startups - Former Y Combinator president has started a new AI safety fund.
LLMS - lmarea.ai has released its beta version of LMArena.
We're excited to invite everyone to a new Beta version of LMArena! š
For months, weāve been poring through community feedback to improve the siteāfixing errors/bugs, improving our UI layout, and more.
To keep supporting the development and continual improvement of this
ā lmarena.ai (formerly lmsys.org) (@lmarena_ai)
5:09 PM ā¢ Apr 17, 2025
Trending in AI ā A company using Cursor had to apologize after the AI agent invented a policy that caused a stir.
Future of Work ā Viral AI art trends are worrying artists about their future.
1. OpenAI Cuts AI Costs With New Flex Processing āļø
OpenAI just rolled out Flex processing, a budget-friendly API option that slashes AI model usage prices by half at the expense of slower responses and occasional downtime, targeting less urgent tasks like data enrichment and model evaluation. This move arrives as AI costs climb and competitors like Google launch cheaper, competitive models, signaling a fierce pricing battle in the AI space.
Flex pricing applies to OpenAIās o3 and o4-mini reasoning models, dropping input token costs dramaticallyāfor example, o3ās input tokens now cost $5 per million instead of $10.
2. Googleās Gemini 2.5 Pro Safety Report Raises Eyebrows š¤
Weeks after launching its most powerful AI model, Gemini 2.5 Pro, Google finally released a safety evaluation reportāonly to face criticism for its lack of detail and transparency. Experts say the report omits key information like findings from Googleās Frontier Safety Framework and doesnāt include tests on all risky capabilities, making it tough to judge the modelās real-world safety.
With promises to regulators on transparency still under scrutiny, this reportās timing and content highlight ongoing tensions between innovation speed and responsible AI deployment.
3. Meta Unveils AI Projects to Boost Machine Perception and Collaboration š
Metaās FAIR team has just dropped five groundbreaking AI tools aimed at making machines see, understand, and interact with the world more like humans, with major upgrades in vision, language, robotics, and social reasoning. Their new Perception Encoder sets a high bar for AI āvision,ā excelling in image and video tasks while enhancing language understanding when paired with large models.
Robotics gets a boost with Meta Locate 3D, which helps robots localize objects from natural language commands in 3D environments, promising smarter human-robot teamwork.
4. Congress Pushes Back on Unauthorized AI Oversight in Government āļø
House Democrats are raising alarms over the Elon Musk-led Department of Government Efficiencyās (DOGE) use of AI tools to monitor federal employees and manage bureaucracy, urging the Trump administration to halt unauthorized AI deployments. The concern centers on security risks, privacy issues, and potential conflicts of interest involving Muskās startup xAI, spotlighting the need for formal approval and legal compliance before adopting such technologies.
This comes amid a broader governmental push to streamline AI use while managing its risks, making it a critical moment for how AI reshapes public sector jobs and data privacy.
5. OpenAIās Stargate Eyes Global AI Infrastructure Expansion š
OpenAIās massive $500 billion Stargate project, initially focused on building AI data centers across the U.S., is now reportedly considering moves into the U.K., Germany, and France.
While still primarily U.S.-centered and in the midst of raising its first $100 billion, this shift signals growing ambitions to shape global AI infrastructure. SoftBank is expected to contribute tens of billions in funding, highlighting high-stakes investment in AIās physical backbone.
6. Intel Tightens AI Chip Sales to China Amid Export Controls ā
In a move signaling tightening U.S. tech export rules, Intel has informed Chinese customers that advanced AI processors with high data bandwidth will now require export licenses, following similar restrictions on NVIDIAās AI chips. This latest development, reported by the Financial Times, comes just a day after Nvidia warned of a $5.5 billion hit due to these limits, highlighting growing trade tensions impacting global chipmakers.
Intelās new CEO Lip-Bu Tan is steering the company through these choppy waters as semiconductor stocks face pressure from shifting policies under the Trump administration.
š¦¾How You Can Leverage:
Sprinkling AI on your decade-old business strategy?
That ain't it.
Your 2010 business playbook with AI features bolted on is doomed to fail. Full stop.
That's what Aishwarya Srinivasan, Head of AI Developer Relations at Fireworks AI, told us during our latest episode of Everyday AI.
She broke down why traditional companies can't just "insert some AI at the end" of their existing workflows and expect magic.
Remember when companies could take YEARS to adopt new tech like cloud or mobile?
Those days are gone forever.
Hereās what ya need to know. š
1 ā The real difference between AI then vs. now š„
Ash has been in machine learning since before the whole generative AI craze exploded. What's changed most dramatically in just the last three years?
Text summarization that once required specialized knowledge of GPT-2 models and complex implementations can now be handled by literally anyone with an internet connection.
But here's the mistake 99% of companies are making right now. They read about some shiny AI tool and desperately search for a place to use it in their business.
WRONG DIRECTION.
Ash explained that smart companies reverse the equation. They start by examining where their employees actually spend time. Which tasks require genuine human critical thinking? Which ones are mundane, repeatable, and ripe for AI assistance?
Your first step isn't finding cool AI tools. It's finding your bottlenecks.
Try this: Document how your team currently spends their time. Be ruthlessly honest.
Create two simple categories: "Critical Human Judgment Required" and "Repeatable/Low-Risk Tasks." Then focus your AI implementation efforts EXCLUSIVELY on that second category first. You'll get faster ROI and build organizational confidence in the tech.
2 ā How to break your brainās business-as-usual autopilot š§āāļø
The biggest obstacle to adopting an AI mindset isn't technology or budget. It's inertia.
We're creatures of habit, comfortably running on autopilot with our trusted workflows. Breaking that pattern requires conscious effort.
Ash shared a killer example from her own work. She wanted to create educational comics about AI topics (returning to her childhood love of art). Her first attempt took a grueling 8-9 hours. Her second try was slightly faster at 5-6 hours.
But after embracing AI tools? She created a quantum computing comic in under 30 minutes.
That's not a small efficiency boost. That's a fundamental transformation in what's possible.
And no, this isn't about replacing jobs. It's about creating space for new opportunities. When your team's mundane tasks shrink from consuming 80% of their day to 20%, they don't become obsolete. T
hey become free to tackle challenges you couldn't previously afford to address.
Try this:
Pick ONE routine workflow that eats at least 2 hours of your week. Don't just think about automating it.
Completely reimagine it as if you were designing it from scratch today, with all AI tools at your disposal. What steps would you eliminate entirely?
What parts actually need human attention? Go nuclear on your assumptions.
1 ā The simplest advice that actually works š£ļø
You're not alone.
Ash's advice is refreshingly straightforward: try AI tools yourself. Personally. Not as a strategic corporate initiative. Just as a human with problems to solve.
See an interesting AI tool mentioned on LinkedIn or Twitter? Give it a spin. The overwhelm melts away once you start experimenting directly.
Even Ash's mother now uses ChatGPT to fact-check news stories she reads online. If she can develop an AI mindset in her daily life, your organization certainly can.
The more you experiment, the faster you'll develop an intuition for separating genuine value from empty hype.
Try this:
Block 30 minutes on your calendar each week labeled "AI play time." During this sacred slot, test ONE new AI tool you've heard about.
No agenda, no ROI analysis, just pure exploration. Keep a simple note about what impressed you and what didn't. After a month, you'll have a personalized toolkit and a much clearer sense of what's actually possible versus what's just marketing fluff.
Remember: Companies viewing AI as just another tech implementation are already toast. The winners have developed an AI mindset at their core.
Reply