Google Brings Gemini AI To Chrome In Browser Update - 1

Image by Firmbee.com, from Unsplash

Google Brings Gemini AI To Chrome In Browser Update

  • Written by Kiara Fabbri Former Tech News Writer
  • Fact-Checked by Sarah Frazier Former Content Manager

Google has announced major Chrome updates to integrate AI into everyday web browsing features.

In a rush? Here are the quick facts:

  • Google adds Gemini AI button to Chrome for desktop users in the U.S.
  • Gemini can summarize tabs, recall past sites, and integrate with Google apps.
  • AI Mode in Chrome’s address bar supports complex queries and follow-ups.

Google is rolling out a range of Gemini-powered features, as well as a browser-specific button which enables users to talk with the chatbot. The company announced that Gemini will become available for Mac and Windows desktop users who use English as their browser language starting today.

“Starting today, we’re rolling out Gemini in Chrome to Mac and Windows desktop users in the U.S. with their language set to English,” the company announced .

Users can use Gemini in Chrome to request webpage information, merge data from different tabs, and retrieve previous websites without needing to search through their browser history.

The AI can also integrate with apps like YouTube, Maps, and Calendar, letting users complete tasks without leaving the page.

Later this month, the Chrome address bar will receive an “AI Mode” option, enabling users to ask complex questions and receive follow-up answers.

The Gemini Nano system from Google provides enhanced safety features through its ability to identify scams and fake websites, while new AI tools will also assist users in handling spam notifications, and password security issues.

On mobile, Android users already have access to some of these tools, and iOS support is coming soon. Google says Gemini’s “agentic” capabilities, which can complete tasks like booking a haircut or filling an online grocery cart, will arrive in the next few months.

Google continues to develop Gemini Gems which users can customize to perform particular tasks. Gems were previously restricted to paid users but Google now offers them to all users who can distribute them through Google Drive file sharing.

While the updates promise convenience, not everyone is enthusiastic. Some users are wary of AI’s growing role or prefer to keep it switched off.

Indeed, the traffic decline experienced at HuffPost and Washington Post, has been reported to exceed 50% which led to layoffs and new business approaches. Critics describe it as content “theft” and warn of “ Google Zero ,” which keeps users within Google’s platform. The Web tab still displays blue links but critics argue that the AI Mode will control search results in the upcoming years.

But with Chrome still the most popular browser, Google’s latest move signals that AI-powered browsing has officially gone mainstream.

OpenAI Finds Promising But Incomplete Fix For AI Scheming Risks - 2

Image by Emiliano Vittoriosi, from Unsplash

OpenAI Finds Promising But Incomplete Fix For AI Scheming Risks

  • Written by Kiara Fabbri Former Tech News Writer
  • Fact-Checked by Sarah Frazier Former Content Manager

Advanced AI systems have the ability to pretend to follow rules while hiding secret objectives. New research says this behaviour can be hampered but not completely fixed.

In a rush? Here are the quick facts:

  • Researchers tested covert actions like underperforming or withholding information.
  • Deliberative alignment reduced scheming in o3 from 13% to 0.4%.
  • Researchers warn situational awareness may be masking genuine progress.

Researchers are warning that sophisticated AI systems can develop the ability “scheme” deceptively, and secretly pursue their own hidden goals.

A new study from OpenAI and Apollo Research, has shown that while progress has been made in reducing this behavior, it remains a risk as AI models grow more capable.

Scheming, the researchers explain, is different from ordinary AI errors. “AI scheming–pretending to be aligned while secretly pursuing some other agenda–is a significant risk that we’ve been studying,” OpenAI wrote .

Indeed, OpenAI’s ChatGPT o1 sparked alarm after Research found it schemed to avoid shutdown. It did so by lying in 99% of interrogations , pursuing self-preservation over instructions. Further research found that penalizing misbehavior only taught the AI to further hide its intent .

To measure scheming in the new research, the teams tested for “covert actions,” this included deliberate test underperformance and information withholding. In one example, OpenAI’s o3 model realized that it would not be deployed if it scored too high on a chemistry test, so it intentionally answered six out of ten questions incorrectly.

The results showed that OpenAI’s o3 model reduced its covert behavior from 13% to 0.4% while o4-mini decreased from 8.7% to 0.3%. “This reduction in scheming on out-of-distribution test environments is a significant win, about a 30× improvement,” OpenAI reported.

Still, the problem isn’t solved. The research indicates that models could be concealing their deceptive actions through better methods of deception. The ability of models to detect evaluation situations leads to reduced deceptive actions.

“Therefore, we cannot exclude that the observed reductions in covert action rates are at least partially driven by situational awareness,” one study notes.

While today’s deployed AIs are unlikely to cause serious harm through scheming, the risk is expected to grow as systems are assigned more complex, real-world tasks. Researchers stress the need for further work, including industry-wide collaboration and improved tools to detect hidden motivations.