• Everyday AI
  • Posts
  • Ep 629: Google’s surprise release: Will Gemini Enterprise Compete with ChatGPT and Microsoft Copilot?

Ep 629: Google’s surprise release: Will Gemini Enterprise Compete with ChatGPT and Microsoft Copilot?

Inside Google Gemini Enterprise, U.S. Senate passes bill prioritizing U.S. chip production, OpenAI urges EU to halt Big Tech competitors, Copilot gets file creation features and more.

Author

Everyday AI
October 10, 2025

Outsmart The Future

Today in Everyday AI
8 minute read

🎙 Daily Podcast Episode: Google just released a new offering for enterprises: Google Gemini Enterprise. We break down what it is and how it works. Give it a watch/read/listen.

🕵️‍♂️ Fresh Finds: Apple looking for new head of AI, Mark Cuban’s genius Sora play, why Trevor Noah says OpenAI’s Sora will be disastrous. Read on for Fresh Finds.

🗞 Byte Sized Daily AI News: U.S. Senate passes bill prioritizing U.S. chip production, OpenAI urges EU to halt Big Tech competitors, Copilot gets file creation features and more. Read on for Byte Sized News.

💪 Leverage AI: Can Google Gemini Enterprise compete with ChatGPT and Microsoft Copilot? Keep reading for that!

↩️ Don’t miss out: Did you miss our last newsletter? We covered: Our 7 Steps to ROI with AI, Ex-Google CEO says AI could be deadly, Gemini to Challenge OpenAI, early signs of AI bubble and more. Check it here!

Ep 629: Google’s surprise release: Will Gemini Enterprise Compete with ChatGPT and Microsoft Copilot?

Will it be a ChatGPT or Microsoft Copilot killer? 

We got our hands on a version of the newest release and will break down everything you need to know, including the ONE feature that could ultimately set Gemini Enterprise apart. 

Google’s surprise release: Will Gemini Enterprise Compete with ChatGPT and Microsoft Copilot?

Also on the pod today:

• Data grounding in Gemini 🤓
• Gemini’s contextual data access 🗃️
• Gemini catching ChatGPT? 🏃‍♂️

It’ll be worth your 39 minutes:

Listen on our site:

Click to listen

Subscribe and listen on your favorite podcast platform

Listen on:

Here’s our favorite AI finds from across the web:

New AI Tool Spotlight – Attrove AI summarizes your most important messages across platforms, RedPill AI gives you encrypted access to 200+ AI models, Lyra is an AI-native meeting platform

AI Safety — Anthropic, the UK AI Security Institute, and The Alan Turing Institute found just 250 poisoned documents can backdoor LLMs of many sizes.

AI Projects — Sam Altman, Shaq and AI. What a combo. Here’s why they’re making headlines.

AI Platform updates — Zendesk unveiled major AI upgrades — from autonomous voice agents and video support to next-gen analytics.

AI and Media — Trevor Noah warns AI video tools like Sora could be “disastrous” Here’s why.

ChatGPT Connectors — Pro users got some new Connectors in ChatGPT.

Apple and AI — Apple eyes new AI chief amid Siri setbacks and reshuffle

LLM Safety — OpenAI says GPT-5 shows 30% less political bias than past models, especially on neutral prompts

Celebrities on Sora — See how Mark Cuban is letting anyone use his Cameo on Sora, yet in a super smart way.

 

1. U.S. Senate Bill to Prioritize U.S. AI Chips 💾

The U.S. Senate has passed the bipartisan GAIN AI measure within the NDAA, directing chipmakers like Nvidia and AMD to prioritize orders for American customers over exports, with an explicit focus on limiting shipments to China and its allies, according to Bloomberg.

Sponsors Senators Elizabeth Warren and Jim Banks say the bill aims to keep startups and small firms from losing out to overseas buyers while strengthening U.S. competitiveness in AI and advanced tech. Nvidia pushed back, calling the policy unnecessary and arguing its H20 exports do not affect supplies of H100, H200, or Blackwell chips. If enacted by the House, this could speed domestic access to high-end accelerators for growing teams and early-stage companies

2. OpenAI Urges EU to Keep Big Tech Competitors in Check ✅

In a timely move, OpenAI met with European Commission officials on September 24 to warn that tech giants like Apple, Microsoft, and Google are gatekeeping the AI market by controlling operating systems and app stores, according to a new disclosure.

The company wants regulators to act fast so users aren’t trapped in closed ecosystems that favor in-house AI products, making life harder for startups and smaller players. Ironically, OpenAI has just inked deals with both Microsoft and Apple, highlighting its complicated dance between partnership and pushback.

3. Sora Rockets Past 1M Downloads In Under Five Days 🚀

OpenAI’s Sora hit one million downloads in less than five days, topping the US App Store and outpacing ChatGPT’s launch pace, according to reporting by BBC and Axios.

The text-to-video app’s surge is fueling social feeds with ten-second clips, but it is also drawing criticism for deepfakes of deceased public figures and copyrighted characters, with OpenAI citing “strong free speech interests” while allowing opt-outs for some recently deceased. Sam Altman says the company is “learning quickly” and plans more granular controls and future revenue-sharing, yet ongoing lawsuits across the industry signal that legal and licensing questions could shape how these tools evolve.

4. Copilot for Windows gets smarter with document creation and inbox connectors 🪟

Microsoft is rolling out an updated Copilot app to Windows Insiders that can generate Word, Excel, PowerPoint, and PDF files directly from a chat, with an export button appearing for longer responses. The app now supports opt-in connectors to Outlook, Gmail, Google Drive, Google Calendar, Google Contacts, and OneDrive, letting users ask Copilot to surface invoices or pull contact details from their email.

This is a timely boost that turns Copilot into a practical workflow hub, cutting the friction between brainstorming and producing shareable docs while making inbox searches and account data retrieval faster.

5. Meta Orders Workers to Build ‘5X Faster’ With AI 🏎️

According to 404 Media, a Meta executive overseeing metaverse products told employees to use AI to “go 5X faster,” as Mark Zuckerberg says most of the company’s code will be written by AI within 18 months.

The timing signals a hard pivot toward AI-first workflows inside Meta, making automation the default in coding and product development. If this sticks, builders and startups may feel pressure to adopt AI tools to match pace, reshaping hiring, training, and how features ship.

🦾How You Can Leverage:

Some of Google's first AI launches weren't home runs.

Bard was rough. Early Gemini was clunky. Workspace AI integration was wonky.

But Google's Gemini Enterprise that it just released yesterday?

Unlike some of Google's earlier first versions of AI products, its Gemini Enterprise product feels and performs like it could quickly become the defacto AI Operating System for large teams.

Google just became a legitimate top-two enterprise player when they weren't 19 months ago.

But, can they catch Microsoft Copilot and ChatGPT? And how does it even work?

(Glad you rhetorically asked.)

Make sure to watch/listen to today's show, and let's dive into what you need to know.

1 – Grounding Might Actually Solve The Hallucination Problem 🔥

Most enterprise AI outputs blend training data into responses even when you connect company files.

They mean well. They just can't help themselves.

Grounding in Gemini Enterprise works differently. It forces the AI to filter everything through your actual company information before responding.

When you ask about internal projects using vague shorthand your team uses, it analyzes patterns across connected sources to decode meaning. Calendar events get triangulated against email threads against document mentions automatically.

The citation trail shows exactly which emails, calendar events, and documents informed each claim.

Regular enterprise AI does semantic search across files one by one. Grounding federates access across your entire data ecosystem while respecting role-based permissions.

So what happens when you upload documents into most LLMs and start asking questions?

They confidently pull in adjacent topics using training data whether you wanted it or not.

Could this (small but huge) Gemini Enterprise feature be the thing that finally makes enterprise AI trustworthy enough for critical business decisions?

Maybe.

Try This

Connect Gmail and Drive to Gemini Business if you're under their 300-seat limit for that tier.

Test it with internal abbreviations your team uses daily without explaining what they mean, then ask about project status across meetings and emails using that exact shorthand.

Compare the responses to what ChatGPT gives you with identical data connections and see if it's mixing in training data you can't verify.

2 – The Pricing Gap Is Kinda Wild ⚡

ChatGPT Enterprise costs $60.

At 5,000 employees, you're looking at $1.8 million in annual savings while getting Google’s Gemini 2.5 models (and soon Gemini 3 models) that consistently top benchmark leaderboards like LMsys Arena.

Gemini Business at $21 monthly gives teams under 300 seats similar grounding capabilities without needing enterprise contracts.

Both tiers connect to Microsoft 365 natively.

Think about that for a second. You can ground AI responses in Outlook, OneDrive, SharePoint, and Teams data while using Google's models instead of being locked into Copilot's performance.

That's gotta sting a bit for Microsoft.

Try This

Pull up your current ChatGPT Enterprise or Microsoft Copilot contract and calculate the actual annual cost per seat including all those fun hidden fees.

Compare that to Gemini Enterprise at $30 monthly, then model what the savings look like at 100 seats, 500 seats, and 2,000 seats to see how quickly this compounds.

Consider running a 50-seat pilot on Gemini Business at $21 before you commit to a full enterprise deployment and multi-year contract.

3 – Google Just Entered The Enterprise Chat(bot) Conversation 🚀

About 18 months ago, we said that Google wasn't a serious enterprise AI contender for business teams.

Vertex served technical devs pretty well, as did AI Studio.

But actual business teams implementing AI in early 2024? They split almost entirely between ChatGPT and Microsoft Copilot.

The grounding prevents the hallucination issues that still plague other platforms. The agent builder feels more limited compared to OpenAI's tool that just shipped, but here's the thing: core performance on data-grounded queries probably matters more than fancy features most teams won't actually use.

Google's positioning this as unified workplace AI that consolidates agent building, enterprise search, and governance in one interface.

Will they actually steal meaningful market share from both Microsoft and OpenAI?

Some. Undoubtedly.

Copilot users fed up with access and permissions issues might switch because Gemini benchmarks better on most tests and costs less.

ChatGPT Enterprise customers with more straightforward needs could convert on price alone when they're deploying thousands of seats and doing the annual budget math.

Try This

Create test prompts that require connecting information across email, calendar, and documents using company-specific terminology without explaining any abbreviations.

Run those identical prompts across ChatGPT, Claude, Copilot, and Gemini Business to document which platforms actually ground responses versus which ones confidently hallucinate plausible-sounding details.

Share the comparison results with whoever handles AI procurement decisions at your company, especially if you're currently paying $60 per seat when Google's offering costs half that with potentially superior data grounding.

What are your thoughts?

Hit us with a reply. We’d love to hear.

Reply

or to participate.