Google Gemini and ChatGPT just rolled out some genuinely big upgrades, and these are not tiny quality-of-life tweaks. These updates change how AI fits into everyday work, browsing, research, finance, and automation. If you use Google Chrome, Gemini is becoming far more useful directly inside the browser. If you use ChatGPT, it can now connect to financial accounts for personalized money insights, and business users can build AI agents with surprisingly little effort.
What makes this wave of releases so interesting is that both companies are moving beyond simple chat. They are turning AI into something that can actually act, organize, analyze, and assist across the tools people already use. And that shift matters a lot more than another minor model benchmark.
Google Gemini inside Chrome is finally getting serious
One of the biggest changes is that Gemini is now showing up directly in Google Chrome. That may sound simple on the surface, but the real value is what it enables.
Inside Chrome, Gemini can now work with Skills. Think of Skills as reusable shortcuts or mini workflows that help Gemini perform a specific kind of task quickly. Instead of rewriting the same prompt over and over, you can save a useful instruction once and trigger it whenever you need it.
This is the kind of feature that starts small and becomes massively useful once it becomes part of your routine.
What Gemini Skills actually do
When using Gemini in Chrome, typing a slash lets you pull up specific Skills. Some are built in. Others can be created and customized.
Examples include:
- Explain simply for breaking down complex material
- Action items for extracting tasks from a page
- Translate text for converting content into another language
- Custom writing and research workflows based on your own prompts
A practical example is using “Explain simply” on a financial article. Instead of getting dense language and jargon, Gemini can turn the page into something much easier to understand. That is helpful not just for beginners, but for anyone dealing with technical content at speed.
Custom Skills are where this gets really powerful
The built-in Skills are nice, but the bigger unlock is creating your own.
For example, you could make a custom Skill called Create YouTube Script that tells Gemini to turn the current page into:
- A YouTube title
- A YouTube script
- A YouTube description
- Relevant tags
Once saved, that workflow becomes reusable. Open a page, trigger the Skill, and Gemini handles the conversion. If your job involves content, education, research, marketing, or internal communications, that kind of saved prompt can cut a lot of repeated work.
Another example is a custom Skill for a school or community use case. You could instruct Gemini to take an article or page and turn it into a post for people trying to learn more about AI. That means one source can be repurposed into a different format for a different audience, without starting from scratch every time.
Google also includes an Enhance Prompt option, which improves the wording of your Skill before you save it. That is a smart addition because a lot of people know what they want AI to do, but struggle to phrase it clearly.
You can browse prebuilt Gemini Skills too
Google is also offering a library of Skills you can browse and add. These are grouped into categories such as:
- Fun
- Learning
- Research
- Shopping
- Understanding
- Writing
This matters because it lowers the barrier to entry. You do not have to be a prompt engineer or an AI power user to get value. You can start with Google’s templates, customize them a little, and build from there.
Why Gemini in Chrome matters more than it first appears
Google already has one huge advantage: distribution. Chrome is already where a massive amount of internet activity happens. So when Gemini becomes deeply useful inside Chrome, AI stops being a separate destination and starts becoming part of normal browsing behaviour.
That is a big strategic shift.
There is also a shortcut that makes Gemini easier to access. With Ctrl + G, Gemini can pop up across windows or tabs, making it feel less like a standalone assistant and more like a built-in layer on top of your browsing.
That is exactly where browser AI is heading: less friction, more context, more assistance without constantly switching tools.
Other Chrome upgrades bundled into this release
Google is not only adding Gemini features. Chrome itself is getting more capable, and some of these upgrades remove the need for separate tools.
1. View two pages in one tab
Chrome now allows a split-page style setup where two pages can be viewed inside one tab. For research, comparison shopping, writing, or fact-checking, that is incredibly useful.
Instead of constantly bouncing between tabs, you can keep source material and output side by side.
2. Smarter tab groups
Chrome tab groups are also becoming more useful for organization. You can group tabs based on a purpose or project, such as:
- Vacation planning
- Research
- Running shoes
- Work tasks
That sounds basic, but when paired with Gemini, these grouped browsing sessions become much easier to manage and summarize.
3. PDF annotation and signing
Chrome now supports annotating and signing PDFs directly. That means one less app, one less browser extension, and one less reason to leave your workflow.
When a browser starts handling tasks that previously required extra software, a lot of standalone tools become less necessary. That is why these updates feel bigger than a feature list. They signal a platform bundling more and more capability into one place.
New Gemini models and live assistance options
There is also an upgrade to the Gemini model options available through Ask Gemini. Users can access multiple model variants, including:
- Flash 3.1 Lite
- Flash
- 3.1 Pro
There is also an option for extended thinking. In simple terms, this adds more reasoning depth to responses. Depending on the task, that can improve how Gemini handles more complex requests.
Another notable feature is the ability to go live with Gemini. That allows Gemini to help explain what you are doing in real time, almost like on-demand tech support or guided assistance. If you are stuck in a workflow, troubleshooting a process, or trying to understand a tool, that live support angle could be extremely useful.
The bigger picture is that Google seems to be finally pushing harder on browser-native AI. Given how much surface area Google already controls, this could turn into a very significant shift very quickly.
ChatGPT Finance brings AI into your actual financial life
The other massive update is on the ChatGPT side: Finance.
For users on the Pro plan, ChatGPT can now connect to financial accounts such as:
- Bank accounts
- Credit cards
- Debit cards
- Investment portfolios
According to the rollout details, this is expected to come to Plus shortly as well.
This is a much bigger deal than “AI gives financial advice” headlines make it sound. The real change is that ChatGPT is moving from general financial suggestions to context-aware analysis based on your own data.
What ChatGPT Finance can help with
Once connected, ChatGPT can help with several practical financial tasks.
Goal planning
You can ask something like: help me build a plan to buy a house in my area in the next five years.
Because ChatGPT can see income, debts, balances, and spending patterns, it can give much more grounded guidance. It can factor in things like:
- Credit card balances
- Cash on hand
- Income levels
- Down payment targets
Vacation cost breakdowns
You can ask what your recent vacations actually cost. Instead of rough guesses, it can review the full spending trail, including:
- Lodging
- Flights
- Food
- Shopping
- Travel extras like Wi-Fi
That kind of breakdown is useful because lots of spending is easy to underestimate when it is spread across multiple transactions.
Spending analysis and budgeting
ChatGPT can also chart spending patterns and surface opportunities to save. This is where visual presentation becomes important. Instead of a wall of numbers, the finance interface can organize information into more readable summaries and graphs.
A prompt like “help me save a little more over the next few months” becomes far more useful when the AI can actually see where the money is going and suggest realistic targets.
Scenario planning
One of the more interesting use cases is decision support. For example:
Can I afford to take a lower-paying job if it gives me more flexibility?
That is the kind of question people wrestle with all the time, and it is not purely theoretical. ChatGPT can evaluate it against your real financial picture, including liquid cash, investments, spending, and normal income patterns.
Portfolio risk analysis
You can also ask what the biggest risk in your portfolio is. ChatGPT can review holdings and identify areas such as:
- Concentration risk
- Allocation issues
- Liquidity risk
That does not mean it replaces a financial professional, but it does make portfolio review more accessible and easier to understand.
Subscription cleanup
One of the most practical features might also be the most relatable: reviewing subscriptions and deciding what to cancel.
That is the kind of task people put off because it is annoying, not because it is difficult. AI is particularly good at handling that sort of low-energy administrative work.
How ChatGPT connects to financial accounts
ChatGPT is using Plaid to connect financial accounts. Plaid is widely used for securely linking bank and investment accounts to financial apps and services.
Supported institutions shown include major names such as:
- American Express
- Bank of America
- Charles Schwab
- Robinhood
- E-Trade
That matters because trust and security are the first questions people ask with a feature like this, and reasonably so.
Privacy and control features
There are a few key safeguards worth noting:
- You remain in control of your linked accounts
- You can disconnect accounts from settings or the Finance page
- Financial memories can be deleted through ChatGPT settings
- Temporary chats do not get access to connected financial accounts
The memory controls are especially important. If ChatGPT stores useful financial details for convenience, you can still manage and remove them through the memory settings under personalization.
If you are cautious about sensitive data, that degree of control is essential.
Why the finance feature is such a big shift
Most AI tools have been strongest at summarizing public information or helping with generic tasks. Finance changes that by giving ChatGPT access to private, structured, personal context.
That means the answers become less generic and more actionable.
Without financial data, “how can I save more money?” gets broad advice. With financial data, the answer can be tied to actual spending categories, realistic budget targets, and your personal cash flow. That is a completely different level of usefulness.
It also points to a broader trend: AI is being plugged into the systems that already contain the details of your life and work. That creates more value, but it also raises the importance of permissions, transparency, and memory controls.
ChatGPT Business now includes agents
The final major release is for ChatGPT Business users: agents.
This is one of the more exciting developments because it lowers the difficulty of building useful AI automations inside ChatGPT. Instead of stitching together multiple tools manually, users can start from templates and customize from there.
Agent templates already cover a lot of work
Available templates include roles and workflows such as:
- Chief of staff
- Customer reply drafter
- Data analysis
- Sales assistant
- Contract clause triage
- Customer support
- Industry briefs
- Knowledge search
- Marketing strategy
- Product planning
These agents can connect to tools, access knowledge sources, and perform tasks based on their configured permissions and capabilities.
That means they are not just generating text. They are closer to operational assistants.
What makes these agents useful
After choosing a template, you can customize the agent by changing:
- The tools it can use
- The connectors it can access
- The skills it has
- How it responds
- How often it runs
You can also schedule agents to run hourly or daily, which pushes this well beyond one-off prompting. That is where automation starts to become meaningful for teams.
A customer reply drafter, for example, could help process support work faster. A chief of staff agent could coordinate across tools and summarize important updates. A data analysis agent could reduce repetitive reporting tasks.
This is the kind of release that can save real time if it is set up well.
The bigger pattern behind all of these launches
Google and OpenAI are both moving in the same general direction:
- More context
- More direct access to tools and data
- More reusable automation
- Less friction between asking and doing
Gemini’s Chrome Skills make browser-based AI more practical. ChatGPT Finance makes personal analysis more tailored. ChatGPT agents make workplace automation easier to deploy.
Put together, these upgrades show that the AI race is no longer just about who has the smartest chatbot. It is about who can build the most useful system around the model.
Final thoughts
These releases are wild not because they sound futuristic, but because they are becoming immediately practical. Google is making Gemini far more useful where people already spend their time, inside the browser. OpenAI is pushing ChatGPT into financial analysis and operational automation in a way that feels much more concrete than generic chat.
If this is the direction both platforms are heading, the next phase of AI will be less about asking clever questions and more about building systems that quietly handle real work for you.
If you are experimenting with AI seriously, now is a good time to start thinking in terms of saved workflows, connected data, and repeatable automations. That is where the real leverage is starting to show up.
And if you want to keep pushing further, explore more ways to make ChatGPT and Gemini automate your work, test custom Skills, and start treating these tools like infrastructure instead of novelties.
FAQ
What are Google Gemini Skills in Chrome?
Gemini Skills are reusable prompts or shortcuts inside Google Chrome that let Gemini perform specific tasks, such as simplifying a page, translating text, extracting action items, or running custom workflows you create yourself.
Can I create my own Gemini Skills?
Yes. You can create custom Skills, name them, write the prompt instructions, enhance the prompt if needed, and save the workflow for repeated use later.
What does ChatGPT Finance do?
ChatGPT Finance connects to your financial accounts and uses your actual financial data to help with budgeting, planning, subscription reviews, spending analysis, goal setting, and portfolio-related questions.
Is ChatGPT Finance available to everyone?
It is currently available on the Pro plan, with rollout to Plus expected shortly after. Availability may depend on your account and region.
How does ChatGPT connect to bank and investment accounts?
ChatGPT uses Plaid to securely connect to supported financial institutions, including major banks and investment platforms.
Can I remove my financial data from ChatGPT?
Yes. You can disconnect linked financial accounts and manage or delete stored financial memories from ChatGPT settings. Temporary chats also do not access connected financial accounts.
What are ChatGPT Business agents?
They are configurable AI assistants inside ChatGPT Business that can use tools, connectors, and templates to automate recurring work such as drafting replies, analyzing data, searching knowledge bases, or planning projects.
Why are these Gemini and ChatGPT updates important?
They move AI beyond simple chat. These updates make AI more useful for real tasks by adding saved workflows, live assistance, financial context, and business automation through agents.
Meta description
Google Gemini and ChatGPT just launched huge new features, including Chrome Skills, ChatGPT Finance, and AI agents for business automation.
