New Google Gemini-SQL2 Update is Insane!

Key Takeaways

New Google Gemini SQL 2 update is insane. Google just dropped something that I think most people are completely sleeping on. It's called Gemini SQL 2. And what it does is genuinely one of the most useful things I've seen AI do for businesses this year. Here's what it is in one sentence. You type a question in plain English, Gemini-SQL 2 writes the database query for you, and it runs it, and it works. So, instead of hiring someone who knows how to code SQL, which is this specific database language that most people never learn, you just ask the question like you would ask a person. Which of our members joined in the last 30 days and haven't posted yet? Done. Which tutorial had the most engagement this month? Done. You ask it like a human. It goes and finds the answer. Hey, if we haven't met already, I'm the digital avatar of Julian Goldie, CEO of SEO agency Goldie Agency. Whilst he's helping clients get more leads and customers, I'm here to help you get the latest AI updates. And Julian Goldie reads every comment, so make sure you comment below. Now, here's why that's a big deal. Gemini SQL 2 just scored 80.04% execution accuracy on the Bird benchmark, the number one spot on the single model leaderboard. The previous best, 77.2%. Google jumped three points in a single release. Mark Tech post threads, and Bird isn't some easy test. It covers 12,751 question SQL pairs across 95 databases in 37 professional domains. It doesn't just check if the SQL looks right. It measures whether the SQL actually runs and returns correct results. AI Weekly Mark Tech post. For context, human performance on Bird sits at 92.96%. Gemini SQL 2 is now about 13 points behind humans. That gap is closing fast. AI Weekly. And Google's prior version, Gemini SQL, was at roughly 77.2%. AWSQL SQL was at about 76.5%. Claude Opus from Anthropic was sitting around 70.1%. Gemini SQL 2 is now clearly ahead AI weekly. So Google is number one in the world at turning plain English into working database queries. And they're saying this is going into their data products. Think BigQuery, Looker, the whole enterprise data stack Google showed off at Cloud Next 2026. Explains this changes something fundamental. Right now most businesses are sitting on a mountain of data they can't actually use. Not because the data isn't there, because nobody on the team knows how to write the code to pull it out. You'd need a data analyst or a developer or someone who knows SQL. Most small business owners don't have that. Gemini SQL 2 removes that wall entirely. So let's talk about what this actually looks like in practice. Imagine you run a community like say the AI Profit Boardroom. You've got thousands of members, you've got engagement data, you've got sign-up data, tutorial views, community activity, all of it sitting in a database somewhere. Before Gemini SQL 2, if you wanted to know something like which members signed up this month but haven't attended a single coaching call yet, you'd have to either pay someone to write that query or just not know. Most people just don't know. They're flying blind. With Gemini SQL 2, you type that exact question, it writes the SQL, it runs it, you get the list, then you can go follow up with those people, send them a message, make sure they're getting value. That's not a small thing. That's the kind of insight that changes how you run your business. Here's another one. Which tutorial topics in the AI Profit Boardroom had the highest completion rate in the last 60 days? That question used to require an analyst. Now you just ask it, you get the answer, and then you go make more of what's working. This, which members have been active for more than 90 days and haven't posted in the community map yet? That's a re-engagement list. That's people you could reach out to. That's personalized outreach at scale powered by data you already have pulled out by AI. This is what Gemini SQL 2 makes possible. You stop guessing. You start knowing. And here's the open loop because this gets more interesting the deeper you go. Right now Google hasn't released a public API for Gemini-SQL 2. No model card, no technical report, no timeline for when it goes live outside of Google's own products. So, we don't have the standalone tool yet, explain X. But, here's what you need to know. You don't have to wait for the official release to start doing this because Gemini 3.1 Pro is available right now, and you can already build this workflow yourself using the current Gemini API with a schema-grounded prompt setup. The moment Gemini SQL 2 officially ships as a product, you'll be ready. And knowing Google, the destination is clearly BigQuery, Looker, and Google's enterprise data stack. This is going to be inside tools that millions of businesses already use. It's not a question of if, it's a question of when, explain X. And if you want to stay ahead of tools like this and actually learn how to put them to work in your business when they drop, that's exactly what we do inside the AI Profit Boardroom. We've already built resources around connecting AI to your business data, automating the tedious stuff, and using AI to get more customers and grow faster. The moment Gemini-SQL 2 goes live and accessible, we'll be walking members through exactly how to set it up, what prompts to use, and how to connect it to your specific business. Four live coaching calls every week where you can bring your exact setup and get real answers. Daily tutorials, a 30-day roadmap, 2,800 members already using AI to grow their businesses right now. And the community map so you can find and connect with other AI automation users near you. Link in the description or go to aiprofitboardroom.com. So, let's walk through what this looks like when you build it yourself right now. The basic idea is this, you give the AI your database structure, what tables you have, what columns are in each table, what the data means. You write that out in plain text, then you ask your question, and Gemini turns it into a working SQL query. For the AI Profit Boardroom specifically, you connect it to your member database, set up the schema, define what each table holds, then start asking questions. How many new members joined from YouTube in the last 14 days? Pull that list. See what's working in your content. Which coaching call topics had the most live attendees over the last quarter? Use that to plan future calls. Which members are on the community map and have listed their city as Karachi, Dubai, or London? Use that to help members find each other locally, which is one of the biggest value drivers in the boardroom right now. Every one of those questions becomes a working query instantly. Now, here's where it gets more powerful. Once you have the data, you layer Gemini 3.1 Pro on top of it. You're not just pulling numbers now. You're asking the AI to make sense of them. Here are the last 30 days of tutorial completions and coaching call attendance. Tell me what patterns you see and what I should do about them. That's an AI analyst working on your business, not just a query tool, a thinking partner that has access to your actual data. This is the stack. Gemini SQL 2 pulls the data. Gemini 3.1 Pro interprets it. You make the decision. And this isn't only for big businesses with huge teams. This is for anyone who has any kind of database. Your email list, your course platform, your booking system, your community platform, whatever you use to run your business. If it stores data, you can ask questions about it. The people who win in the next 2 years are going to be the ones who can ask the right questions. Gemini SQL 2 handles the technical part, because when Gemini-SQL 2 ships, and it will ship, those questions become answers in seconds. And if you want to stay ahead of tools like this and actually learn how to put them to work in your business when they drop, that's exactly what we do inside the AI Profit Boardroom. We've already built resources around connecting AI to your business data, automating the tedious stuff, and using AI to get more customers and grow faster. The moment Gemini-SQL 2 goes live and accessible, we'll be walking members through exactly how to set it up, what prompts to use, and how to connect it to your specific business. Four live coaching calls every week where you can bring your exact setup and get real answers. Daily tutorials, a 30-day roadmap, 2,800 members already using AI to grow their businesses right now. And the community map, so you can find and connect with other AI automation users near you. Link in the description or go to AI profit boardroom.com. And if you want all the notes from this video, plus 100 plus AI use cases you can start using immediately, including how to build your own text to SQL setup right now with current Gemini models. Come join the AI success lab. It's free. 67,000 members in there already. Link is in the comments and description. Everything from this video is in there, plus a community of people actually doing this stuff every day.