- Everyday AI
- Posts
- Ep 839: Scheduling AI: how to easily make AI work for you in Claude, Gemini and ChatGPT
Ep 839: Scheduling AI: how to easily make AI work for you in Claude, Gemini and ChatGPT
Grok 4.6 drops with impressive benchmarks, Google expands Gemini connectors, Grok Bot brings competition to Codex and more.
👉 Subscribe Here | 🗣 Hire Us To Speak | 🤝 Partner with Us | 🤖 Grow with GenAI
Outsmart The Future
Today in Everyday AI
7 minute read
🎙 Daily Podcast Episode: Scheduled AI tasks can automate repetitive work before you even start your day. We compare how ChatGPT, Claude, and Gemini handle them. Give today’s show a watch/read/listen.
🕵️♂️ Fresh Finds: Microsoft launched a faster, cheaper coding model, OpenAI added agent syncing, and Google unveiled real-time ASL translation. And more. Read on for Fresh Finds.
🗞 Byte Sized Daily AI News: Grok 4.6 drops with impressive benchmarks, Google expands Gemini connectors, Grok Bot brings competition to Codex and more. Read on for Byte Sized News.
💪 Leverage AI: Scheduled AI tasks can automate research, reports, and recurring workflows. Here’s how ChatGPT, Claude, and Gemini compare. Keep reading for that!
↩️ Don’t miss out: Miss our last newsletter? We covered: Zuckerberg's AI manifesto, NVIDIA's new open agent model, former OpenAI COO is leaving and more. Check it here!
Ep 839: Scheduling AI: how to easily make AI work for you in Claude, Gemini and ChatGPT
Scheduled tasks are a secret weapon. ⚔️
How secret?
They can actually be hard to find and there's not a lot of info out there on how to use them. lolz.
But for many, they can be the stepping stone to the fully autonomous desktop worker. Because for many users who may only be able (or comfortable) to access AI on the web, scheduled tasks provide that proactive, work-done-for-you vibe that AI agents delivered.
But how does it work in Gemini, ChatGPT and Claude?
And what's worth scheduling and automating?
We put AI to work on this Wednesday and find out.
Also on the pod today:
• Scheduled tasks: AI’s secret weapon ⏰
• ChatGPT’s “Work” virtual browser mode 💻
• Claude routines + Zapier triggers 🔗
Listen on our site:
Subscribe and listen on your favorite podcast platform
Listen on:
Here’s our favorite AI finds from across the web:
New AI Tool Spotlight – Tines 3B is One environment to deliver AI agents, apps, and automation, BetterClaw Builds AI Agents in 60 Seconds. No Code, bb can control, customize, and automate itself, laying the groundwork for your own software factory.
MAI-Code-1.1 Flash — GitHub Copilot’s new MAI-Code-1.1-Flash model is faster, smarter, and costs a quarter as much as before.
OpenAI Sync — OpenAI just made it possible to sync your projects, skills, and plugins across Codex, ChatGPT Work, and other AI agents
Tencent Worldclaw — Tencent’s WorldClaw builds explorable 3D worlds from a single text prompt, keeping everything editable and spatially coherent.
Google ASL Model — Google’s new SL2T model lets you sign in ASL straight into your phone, translating gestures to text in real time.
Raindrop Signals 2.0 — Raindrop just dropped Signals 2.0, letting you build powerful binary classifiers from production traces at a fraction of the cost.
Spotify AI — Spotify is rolling out AI Persona badges to flag AI-generated artist profiles, aiming for more transparency on who’s really behind the music.
OpenAI Free for College — OpenAI is about to give US college students a full year of ChatGPT Plus for free, but only if their campus is on the list.
OpenAI Desktop on Linux — OpenAI just dropped a ChatGPT desktop app preview for Linux, now with Codex alongside ChatGPT Work.
1. Google Says Gemini Has Surpassed 1 Billion Monthly Users 😋
Google CEO Sundar Pichai announced Monday that more than 1 billion people now use the Gemini app each month, making it the company’s fastest-growing product and its 14th service to reach that scale.
The milestone underscores how quickly consumer AI has moved from novelty to a mainstream Google product, helped by the company’s enormous global distribution.
2. SpaceXAI Unveils Grok 4.6 With Frontier-Model Claims 🏆
SpaceXAI has released Grok 4.6, pitching it as a stronger model for coding, long-running agents, knowledge work, and visual tasks. In terms of overall intelligence, it is on par with GPT-5.6 Sol and Claude Fable 5, while slightly behind Claude Opus 5.
The release is available through several developer platforms today, with API pricing starting at $2 per million input tokens, signaling a push to compete on both capability and cost.
3. OpenAI Brings Daybreak Cybersecurity Models to Amazon Bedrock 👾
OpenAI has made its Daybreak cybersecurity capabilities available through Amazon Bedrock, putting advanced defensive and authorized security-testing tools inside AWS environments.
Daybreak Blue offers safeguarded access to frontier general-purpose models, while Daybreak Red targets approved vulnerability research, exploit validation, and security testing.
4. IBM and Together AI Plan $240M NVIDIA B300 Inference Cluster for 2027 ⚡
IBM has announced a $240 million, multi-year deal with Together AI to build a large NVIDIA HGX B300 inference cluster on IBM Cloud, expected to be available in the first quarter of 2027.
The project is billed as IBM Cloud’s first dedicated large-scale inference deployment using these systems, aimed at making open-source AI faster and cheaper to run in production.
5. Grok Bot Launches AI Teammates 🤝
Grok Bot is launching in beta today, pitching AI agents that can use their own cloud computers to complete work across inboxes, apps, and websites without constant supervision.
The announcement signals a push beyond chat-based assistants toward software that can carry out multi-step office, sales, and engineering tasks, then return only when human approval is needed.
6. Google Adds 14 New Gemini App Connections 🛠️
Announced today at Made by Google, Gemini is expanding its connected-app options with services including Otter.ai, Wix, OpenTable, Ticketmaster, Pandora, Zocdoc, and others.
The rollout, due over the next few weeks, lets users link these services so Gemini can help coordinate tasks from one chat interface rather than requiring constant app switching. It is Google’s latest move to make Gemini less of a standalone chatbot and more of a central control point for everyday digital errands.
You can wake up tomorrow with a chunk of your work already finished, no desktop agent required. The feature is buried so deep, most people never find it.
How?
Scheduled tasks. They're hiding inside ChatGPT, Claude, and Gemini, and they're the gateway drug to a fully agentic workflow. Describe the work once, pick a time, and it just RUNS while you sleep.
"Sounds like a glorified alarm clock."
Wayyy off.
This might be the most underrated capability in AI right now. We don't say that lightly.
What's actually possible? An agent with its own virtual computer clicking around the web overnight, and routines that fire the second a lead hits your site.
Don't worry, zero code required. This is your scheduling 101 for the nontechnical leader.
On today's Everyday AI, we break down which platform wins, plus the hidden hacks nobody's talking about and what to schedule first.
Class starts now. Let's put AI to work.
1. ChatGPT's Hidden Scheduling Hacks 🔥
ChatGPT won our head-to-head, and not because of anything on the label. The good stuff is hidden.
First hack: skip the Scheduled page. Set up your task conversationally in a normal chat and you unlock the model selector the scheduling page quietly hides.
Why does that matter?
Your task runs while you sleep, so throw the most powerful model at it. No reason to default to whatever you used last.
Second hack: schedule inside the new Work mode and your task gets its own virtual computer and browser.
All that clicking, scrolling, and filtering you do on your favorite sites every morning? An agent can now physically DO it overnight.
One more freebie: open-ended tasks like "watch a competitor's pricing page" just work, no schedule needed.
Try This
Open a normal ChatGPT chat, not the Scheduled page, and describe one repetitive morning task with a time attached. Pick your most powerful model before you hit send.
Tomorrow, the output is waiting before your coffee is. Once that sticks, move it into Work mode so it can browse and click for you.
One scheduled task usually turns into five within a week.
2. Claude Routines Trigger on Anything ⚡
Claude's regular scheduled tasks are solid but straightforward. The real power hides in a Claude Code feature called Routines.
You don't need to be technical. Promise.
Here's the gnarly part: Routines can hand you an API token that acts as a trigger. Plug it into Zapier, and anything across 6,000+ connected apps can kick off an agent instantly.
What does that actually look like?
A million dollar prospect fills out the form on your website, and a Claude agent starts the research legwork before you've even seen the notification.
Every other platform waits for the clock. This one waits for your business.
Try This
Open Claude Code, create a new Routine, and choose the API token trigger. Takes about 30 seconds, and Claude walks you through it.
Then connect one high-value event in Zapier, like a new lead form or a big deal moving stages.
The first time an agent finishes legwork before your laptop is open, you'll rethink what else deserves a trigger.
3. Your First Three Scheduled Tasks 🚀
Skip Gemini's Scheduled Actions for now. They're buried in settings, can't use your connected data, and ours pulled in wrong stock numbers.
Hard pass until Google gives them some love.
So what should you actually schedule? Start by asking your AI what you do repetitively inside it.
Those repeat prompts you're prolly running weekly? Automate them.
Sneaky hack: hand an agent your browser history and ask which sites you visit most. Tie each one to a deliverable, then schedule it.
Then the big one: a daily triage dashboard pulling your mail, calendar, Slack, and KPIs into one start-here view.
The 2024 flex was "improve this email." The 2026 flex is waking up to a dashboard that tells you where to begin.
Try This
Tonight, schedule one task: a morning triage dashboard that pulls your top two or three connectors and flags what needs you first.
Keep it simple on day one, then layer in your CRM and KPIs as it earns your trust.
Check it with your morning coffee for one full week. By Friday, you'll wonder how mornings ever worked without it.






Reply