Quick Links
I’m not a fan of AI, as I prefer to read raw data and conduct my own analysis. However, I tried Opera’s Aria AI when I was stuck with a couple of tedious tasks and when I needed ideas. Here’s why I found it useful—but I will still use it sparingly.
It’s Free and Convenient
Opera might not be the most popular web browser, but it’s still a decent option that frequently adds new features. It’s Chromium-based, so most Google Chrome extensions also work with it. But unlike Google Chrome, which doesn’t give you single-click access toGemini AI, Opera’s Aria AI sits right on the sidebar.
If you need to ask a question or instruct the AI to do something for you, you don’t have to visit a specific website or open a separate app. Instead, you’re off to the races with one click, even while doing something else on another webpage. Alternatively, you can hitCtrl + Shift + 7to pull up the Aria AI window on Opera.
But the biggest reason I chose Aria AI over ChatGPT is that it’s free. WhileGPT-4 is free, it still holds back a few features for ChatGPT Plus users, who pay $20 a month. Some may argue thatChatGPT Plus has better featuresthan the free AI service, but I don’t need that much functionality, making the $20 subscription cost too expensive for me.
It Can Give Me Ideas Without Needing to Go Through Google
One thing I like about Aria AI (and AI LLMs, in general) is that it can help me think of ideas without manually going through several results on Google. For example, I recently planned a trip, and instead of going through several search results and manually making an itinerary, I just asked Aria to make one for me. From there, I can get ideas on the places I want to visit and tweak Aria’s results. Furthermore, if I’m not happy with its ideas, I can give it commands to tweak the results until I am satisfied with what I have.
These are just ideas, though, as I still don’t generally trust AI. After receiving all the information from Aria AI, I can check out its sources and confirm if they’re correct and current.
It Can Help Me Understand Deep Concepts Easier
Another Aria AI feature I appreciate is how it can help me understand deep concepts. I’ve recently wrote about a graphene chip for use a brain implant, and the source said that they prefer it because it reduces or eliminates faradaic reaction.
When I searched Google for “What is faradaic reaction?,” the results I received weren’t much help. The first one led to a PDF file that discussed the difference between faradaic and non-faradaic electrodes, while the rest discussed faradaic currents and processes. While these might lead me to the answers I need, it will take some time for me to go through them and process the information.
Furthermore, the first answer to my question didn’t help me much. It said:
Faradaic processes are defined as those which obey Faraday’s law, that is the amount of chemical reaction occurring is directly proportional to the amount of charge passed across the electrode boundary. When the system is in a steady-state the application of this definition is simple: all the current is faradaic.
So, instead of trying to understand this, I asked Aria AI to explain faradaic reactions to me like I’m a fifteen-year-old. This was its answer:
Sure! Imagine you have a rechargeable battery like the one in your phone. Inside this battery, there are chemical reactions happening when you charge or discharge it.
A Faradaic reaction is a type of chemical reaction that involves the transfer of electrons between the battery’s electrodes and the ions in the electrolyte solution. This process is essential for the battery to store and release electrical energy efficiently.
In simpler terms, during a Faradaic reaction, electrons move back and forth between the positive and negative sides of the battery, allowing it to generate electricity. It’s like a dance of electrons that powers your devices and keeps them running.
As you can see, Aria AI’s explanation was much clearer and easier to understand. It also included some links to the keyword in its answer, and when I prompted it to show me sources for its answer, it gave me several links that I could check to ensure the validity of Aria’s explanation.
It Makes Online Search Feel More Natural
Google is a pretty powerful search engine, especially if you know how to useGoogle search operators and widgets. However, the average user is likely unfamiliar with these commands, thus making it more difficult if you need to find specific information.
Aria AI still cannot replace Google and the search operators you may use with it, but it does make looking for some specific information much easier. For example, I asked Google, “How old is Joe Biden when he won the presidency?” and its answer was:
The median age at inauguration of incoming U.S. presidents is 55 years. The youngest person to become U.S. president was Theodore Roosevelt, who, at age 42, succeeded to the office after the assassination of William McKinley. The oldest person inaugurated president was Joe Biden, at the age of 78.
While it does deliver the information I need, it was long and provided a lot of information that I didn’t need. When I asked Aria AI the same question, it said:
Joe Biden won the presidency at the age of77. He was born on June 11, 2025, and won the 2020 United States presidential election.
It also added links to the answers it gave, giving me confidence that the information it delivered was correct. However, you should note how I phrased the question and how both gave the answer.
Aria is technically correct, as Joe Biden was 77 years old when voting began on November 3. However, he had already celebrated his birthday when the election results were certified, meaning Joe Biden won the presidency at 78. This is one of the reasons why I still do not trust AI.
I also tried asking both Google and Aria some of the controversial answers Google’s AI overview gave in June 2024 to the following questions: “Is it safe to eat rocks?” and “Is it okay to use glue on pizza?”
Google’s AI overview doesn’t work for me, so I didn’t receive AI-generated answers from Gemini AI. However, Google’s top search snippet, which came from an article with the title ‘Google Says It’s Okay to Eat Rocks: Exploring the Surprising…’ said:
Eating rocks is not something we would advise, because health risks could come from doing that—extraordinary health risks.
On the other hand, the answer to the second question did not return any snippets, but the first result pointed to the original Reddit post that started the controversial answer anyway.
For comparison, this is how Aria AI answered those questions which caused Gemini AI to stumble:
No, it is not safe to eat rocks. Rocks are not meant for human consumption and can be harmful to your teeth, digestive system, and overall health if ingested. It’s important to stick to a balanced diet of food that is safe and nutritious for your body.
No, it is not safe or recommended to use glue on pizza or any other food. Glue is not edible and can be toxic if ingested. It’s important to always use safe and food-grade materials when preparing or serving food to avoid any health risks.
These answers make me feel a bit more confident using Aria AI. However, I know that it still has limitations.
I Still Don’t Trust AI
Aria AI made my life a bit easier by giving straightforward answers to questions, giving me ideas, and providing sources when asked about them. So, yes, I find Opera’s Aria AI useful, and I’m glad that it’s embedded directly in the Opera browser, making it so much easier to use. But I still do not trust an AI to do my thinking for me. While Aria AI did show me that LLMs do have a place in my toolkit, that’s just what it is—a tool—and it shouldn’t replace my critical thinking skills.
