#agenticbrowsers — Public Fediverse posts

https://winbuzzer.com/2026/03/04/chrome-gemini-flaw-rogue-extensions-hijack-ai-panel-xcxwbn/

Chrome Gemini Flaw Let Rogue Extensions Hijack Chrome AI Panel

#AI #Google #GoogleChrome #Gemini #GoogleGemini #WebBrowsers #Cybersecurity #BrowserExtensions #SecurityVulnerabilities #SecurityFlaws #Exploits #SecurityPatches #AgenticBrowsers #RogueExtensions

https://winbuzzer.com/2026/03/04/chrome-gemini-flaw-rogue-extensions-hijack-ai-panel-xcxwbn/

Chrome Gemini Flaw Let Rogue Extensions Hijack Chrome AI Panel

#AI #Google #GoogleChrome #Gemini #GoogleGemini #WebBrowsers #Cybersecurity #BrowserExtensions #SecurityVulnerabilities #SecurityFlaws #Exploits #SecurityPatches #AgenticBrowsers #RogueExtensions

https://winbuzzer.com/2026/03/04/chrome-gemini-flaw-rogue-extensions-hijack-ai-panel-xcxwbn/

Chrome Gemini Flaw Let Rogue Extensions Hijack Chrome AI Panel

#AI #Google #GoogleChrome #Gemini #GoogleGemini #WebBrowsers #Cybersecurity #BrowserExtensions #SecurityVulnerabilities #SecurityFlaws #Exploits #SecurityPatches #AgenticBrowsers #RogueExtensions

https://winbuzzer.com/2026/03/04/chrome-gemini-flaw-rogue-extensions-hijack-ai-panel-xcxwbn/

Chrome Gemini Flaw Let Rogue Extensions Hijack Chrome AI Panel

#AI #Google #GoogleChrome #Gemini #GoogleGemini #WebBrowsers #Cybersecurity #BrowserExtensions #SecurityVulnerabilities #SecurityFlaws #Exploits #SecurityPatches #AgenticBrowsers #RogueExtensions

https://winbuzzer.com/2026/03/04/chrome-gemini-flaw-rogue-extensions-hijack-ai-panel-xcxwbn/

Chrome Gemini Flaw Let Rogue Extensions Hijack Chrome AI Panel

#AI #Google #GoogleChrome #Gemini #GoogleGemini #WebBrowsers #Cybersecurity #BrowserExtensions #SecurityVulnerabilities #SecurityFlaws #Exploits #SecurityPatches #AgenticBrowsers #RogueExtensions

Tim Berners‑Lee pushes back on Sundar Pichai’s warning that AI will wreck the web. He argues generative AI can enrich the World Wide Web, not erase it, and even hints at “agentic browsers” like OpenAI’s Atlas shaping Chrome’s future. Read how the web’s creator sees AI’s role. #TimBernersLee #SundarPichai #GenerativeAI #AgenticBrowsers

🔗 https://aidailypost.com/news/tim-bernerslee-says-ai-wont-destroy-web-despite-pichais-comment

Tim Berners‑Lee pushes back on Sundar Pichai’s warning that AI will wreck the web. He argues generative AI can enrich the World Wide Web, not erase it, and even hints at “agentic browsers” like OpenAI’s Atlas shaping Chrome’s future. Read how the web’s creator sees AI’s role. #TimBernersLee #SundarPichai #GenerativeAI #AgenticBrowsers

🔗 https://aidailypost.com/news/tim-bernerslee-says-ai-wont-destroy-web-despite-pichais-comment

Tim Berners‑Lee pushes back on Sundar Pichai’s warning that AI will wreck the web. He argues generative AI can enrich the World Wide Web, not erase it, and even hints at “agentic browsers” like OpenAI’s Atlas shaping Chrome’s future. Read how the web’s creator sees AI’s role. #TimBernersLee #SundarPichai #GenerativeAI #AgenticBrowsers

🔗 https://aidailypost.com/news/tim-bernerslee-says-ai-wont-destroy-web-despite-pichais-comment

Tim Berners‑Lee pushes back on Sundar Pichai’s warning that AI will wreck the web. He argues generative AI can enrich the World Wide Web, not erase it, and even hints at “agentic browsers” like OpenAI’s Atlas shaping Chrome’s future. Read how the web’s creator sees AI’s role. #TimBernersLee #SundarPichai #GenerativeAI #AgenticBrowsers

🔗 https://aidailypost.com/news/tim-bernerslee-says-ai-wont-destroy-web-despite-pichais-comment

"AI is a Liability | Weekly News Roundup" 👀👏

https://friendica.world/display/84b6ef2b-1969-0529-1467-645489192884

"AI is a Liability | Weekly News Roundup" 👀👏

https://friendica.world/display/84b6ef2b-1969-0529-1467-645489192884

"AI is a Liability | Weekly News Roundup" 👀👏

https://friendica.world/display/84b6ef2b-1969-0529-1467-645489192884

"AI is a Liability | Weekly News Roundup" 👀👏

https://friendica.world/display/84b6ef2b-1969-0529-1467-645489192884

Some sobering considerations...

Is ChatGPT Atlas safe? What you should know before you download it https://ino.to/wJkmaOo

#AgenticAI #AgenticBrowsers

Some sobering considerations...

Is ChatGPT Atlas safe? What you should know before you download it https://ino.to/wJkmaOo

#AgenticAI #AgenticBrowsers

Some sobering considerations...

Is ChatGPT Atlas safe? What you should know before you download it https://ino.to/wJkmaOo

#AgenticAI #AgenticBrowsers

Some sobering considerations...

Is ChatGPT Atlas safe? What you should know before you download it https://ino.to/wJkmaOo

#AgenticAI #AgenticBrowsers

Some sobering considerations...

Is ChatGPT Atlas safe? What you should know before you download it https://ino.to/wJkmaOo

#AgenticAI #AgenticBrowsers

Question for (people who like the work of) @pluralistic : will the growth of autonomous/agentic browsers improve interoperability?

Let's say I wanted to mirror my Mastodon posts to a Twitter account in a way that is indistinguishable from a human user. Could #AgenticBrowsers help with that?

Similarly, it could make it easier (for e.g. #Nitter, or for Twitter -> Mastodon bridges) to bypass API restrictions?

It's not that I want #Twitter in my life, but that bridges can help people leave?

Question for (people who like the work of) @pluralistic : will the growth of autonomous/agentic browsers improve interoperability?

Let's say I wanted to mirror my Mastodon posts to a Twitter account in a way that is indistinguishable from a human user. Could #AgenticBrowsers help with that?

Similarly, it could make it easier (for e.g. #Nitter, or for Twitter -> Mastodon bridges) to bypass API restrictions?

It's not that I want #Twitter in my life, but that bridges can help people leave?

Question for (people who like the work of) @pluralistic : will the growth of autonomous/agentic browsers improve interoperability?

Let's say I wanted to mirror my Mastodon posts to a Twitter account in a way that is indistinguishable from a human user. Could #AgenticBrowsers help with that?

Similarly, it could make it easier (for e.g. #Nitter, or for Twitter -> Mastodon bridges) to bypass API restrictions?

It's not that I want #Twitter in my life, but that bridges can help people leave?

Question for (people who like the work of) @pluralistic : will the growth of autonomous/agentic browsers improve interoperability?

Let's say I wanted to mirror my Mastodon posts to a Twitter account in a way that is indistinguishable from a human user. Could #AgenticBrowsers help with that?

Similarly, it could make it easier (for e.g. #Nitter, or for Twitter -> Mastodon bridges) to bypass API restrictions?

It's not that I want #Twitter in my life, but that bridges can help people leave?

Question for (people who like the work of) @pluralistic : will the growth of autonomous/agentic browsers improve interoperability?

Let's say I wanted to mirror my Mastodon posts to a Twitter account in a way that is indistinguishable from a human user. Could #AgenticBrowsers help with that?

Similarly, it could make it easier (for e.g. #Nitter, or for Twitter -> Mastodon bridges) to bypass API restrictions?

It's not that I want #Twitter in my life, but that bridges can help people leave?

This article from Brave's security team highlights a significant security risk in agentic browsers, where they can be tricked into doing things such as stealing credentials. They demonstrate a proof of concept in the article.

Agentic browsers have a built-in AI assistant that can browse the web and perform tasks on your behalf. While they speak about Perplexity's Comet browser, know that Edge with Co-pilot has some of these same vulnerabilities, which have been demonstrated. But don't worry. Microsoft says they solved it. There are many agentic browsers in development.

The article details a vulnerability called indirect prompt injection. This attack allows a bad actor to hide malicious instructions on a webpage that the AI will follow as if they were legitimate commands.

Traditional browsers like Firefox that are designed for human use only and lack these autonomous AI capabilities are not vulnerable to this specific attack.

https://brave.com/blog/comet-prompt-injection/

#Brave #AI #LLM #CyberSecurity #Agentic #AgenticBrowsers #Copilot

This article from Brave's security team highlights a significant security risk in agentic browsers, where they can be tricked into doing things such as stealing credentials. They demonstrate a proof of concept in the article.

Agentic browsers have a built-in AI assistant that can browse the web and perform tasks on your behalf. While they speak about Perplexity's Comet browser, know that Edge with Co-pilot has some of these same vulnerabilities, which have been demonstrated. But don't worry. Microsoft says they solved it. There are many agentic browsers in development.

The article details a vulnerability called indirect prompt injection. This attack allows a bad actor to hide malicious instructions on a webpage that the AI will follow as if they were legitimate commands.

Traditional browsers like Firefox that are designed for human use only and lack these autonomous AI capabilities are not vulnerable to this specific attack.

https://brave.com/blog/comet-prompt-injection/

#Brave #AI #LLM #CyberSecurity #Agentic #AgenticBrowsers #Copilot

This article from Brave's security team highlights a significant security risk in agentic browsers, where they can be tricked into doing things such as stealing credentials. They demonstrate a proof of concept in the article.

Agentic browsers have a built-in AI assistant that can browse the web and perform tasks on your behalf. While they speak about Perplexity's Comet browser, know that Edge with Co-pilot has some of these same vulnerabilities, which have been demonstrated. But don't worry. Microsoft says they solved it. There are many agentic browsers in development.

The article details a vulnerability called indirect prompt injection. This attack allows a bad actor to hide malicious instructions on a webpage that the AI will follow as if they were legitimate commands.

Traditional browsers like Firefox that are designed for human use only and lack these autonomous AI capabilities are not vulnerable to this specific attack.

https://brave.com/blog/comet-prompt-injection/

#Brave #AI #LLM #CyberSecurity #Agentic #AgenticBrowsers #Copilot

This article from Brave's security team highlights a significant security risk in agentic browsers, where they can be tricked into doing things such as stealing credentials. They demonstrate a proof of concept in the article.

Agentic browsers have a built-in AI assistant that can browse the web and perform tasks on your behalf. While they speak about Perplexity's Comet browser, know that Edge with Co-pilot has some of these same vulnerabilities, which have been demonstrated. But don't worry. Microsoft says they solved it. There are many agentic browsers in development.

The article details a vulnerability called indirect prompt injection. This attack allows a bad actor to hide malicious instructions on a webpage that the AI will follow as if they were legitimate commands.

Traditional browsers like Firefox that are designed for human use only and lack these autonomous AI capabilities are not vulnerable to this specific attack.

https://brave.com/blog/comet-prompt-injection/

#Brave #AI #LLM #CyberSecurity #Agentic #AgenticBrowsers #Copilot

This article from Brave's security team highlights a significant security risk in agentic browsers, where they can be tricked into doing things such as stealing credentials. They demonstrate a proof of concept in the article.

Agentic browsers have a built-in AI assistant that can browse the web and perform tasks on your behalf. While they speak about Perplexity's Comet browser, know that Edge with Co-pilot has some of these same vulnerabilities, which have been demonstrated. But don't worry. Microsoft says they solved it. There are many agentic browsers in development.

The article details a vulnerability called indirect prompt injection. This attack allows a bad actor to hide malicious instructions on a webpage that the AI will follow as if they were legitimate commands.

Traditional browsers like Firefox that are designed for human use only and lack these autonomous AI capabilities are not vulnerable to this specific attack.

https://brave.com/blog/comet-prompt-injection/

#Brave #AI #LLM #CyberSecurity #Agentic #AgenticBrowsers #Copilot

Brave finds worrying vulnerabiities in Perplexity's Comet, but likely it exists elsewhere

https://feddit.online/post/988556

Brave finds worrying vulnerabiities in Perplexity's Comet, but likely it exists elsewhere

https://feddit.online/post/988556