Google just made Gemini Spark much more powerful 🔥 Spark can now use Chrome to handle complex task. It can use your logged-in accounts and saved passwords, while handing sensitive actions like payments back to you. Video credit: Google Blog

Just tried it, its not control my chrome, refuses to do so.

Slow rollout may be..

yeah true, tried the other stuff too. Not super impressive. this is basically gemini connected with all mcp severs they have, isn't it? am I missing something?

I don't think it uses any mcp for now.

It can’t generate even an image 😬😬

just a shame Europeans will have to wait ages to be able to use this

Very helpful I use it

This is nice, I prefer watching the AI in my browser. I learn a lot and I also stop it from doing stupid things sometimes. I'll check this out

This is a smart compromise for security. 😊

Bro, it gives up after few requests only

payments are where spark suddenly gets polite

I believe this accentuates the shift from prediction to secure interaction. Gemini Spark browsing Chrome natively with secure handoffs is huge. Do we in MENA have stable policies and infrastructure to support this? UAE is paving the way.

Interesting

letting an agent touch logged-in sessions is the part that'd make me nervous, not the payment gate

browser control is the real unlock here, not more tokens. the hard part is safe state handling: auth, session reuse, and strict human handoff for irreversible actions. if google nails that, spark becomes genuinely useful.

Can anyone tell me, is Gemini Spark in conversation mode (listening to you) supposed to respond back with voice? On my Mac I have it, and tried it, and it's damn confusing it doesn't say anything and response seems stuffed into that little pill window at the bottom.

Gemini Spark still has a long way to go to develop.

wow! just tested it and it works this is way better than the remote browser

Coming to India when?

handing payments back to you is a smart line to draw