Image by Pawel Czerwinski, from Unsplash
Windsurf AI Talent Joins Google DeepMind, Sidestepping OpenAI-Microsoft Deal
- Written by Kiara Fabbri Former Tech News Writer
- Fact-Checked by Sarah Frazier Former Content Manager
After OpenAI rejected its $3B acquisition offer, Google acquired Windsurf’s leading AI experts to boost its DeepMind operations in Silicon Valley’s AI competition.
In a rush? Here are the quick facts:
- OpenAI tried to buy Windsurf for $3 billion but talks collapsed.
- Windsurf refused to share sensitive data with Microsoft, OpenAI’s investor.
- Google licensed Windsurf’s tech instead of acquiring the company outright.
Google has hired the top minds behind Windsurf, a start-up that created an AI tool for writing code, adding new strength to its AI division, DeepMind, as first reported by The New York Times . The move is part of Silicon Valley’s growing rush to lock down the best artificial intelligence talent.
CEO Varun Mohan, co-founder Douglas Chen, and several Windsurf researchers now work at Google DeepMind. Google also paid for a license to use Windsurf’s technology, though the amount wasn’t disclosed, the Times noted.
“We’re excited to welcome some top A.I. coding talent from Windsurf’s team to Google DeepMind to advance our work in agentic coding,” Google said in a statement, as reported by the Times.
OpenAI had tried to buy Windsurf in a $3 billion deal just weeks ago. The deal collapsed after Windsurf refused to share sensitive product data with Microsoft, OpenAI’s biggest investor. The Times reports that Microsoft required access to this data as a condition.
Once Windsurf’s exclusivity agreement with OpenAI ended, Google stepped in. CEO Sundar Pichai and DeepMind’s Demis Hassabis offered a simpler deal with fewer complications, says the Times.
This isn’t the first time Google has used licensing and hiring to gain AI talent. The Times reports that in 2023, it paid $3 billion to license Character.AI’s technology and also hired its founders, former Google employees.
Rival companies like Meta and Google have reportedly offered compensation packages as high as $100 million to attract top talent. Windsurf’s business chief, Jeff Wang, will now serve as interim CEO. OpenAI and Microsoft declined to comment further, as reported by the Times.
The FTC and DOJ usually watch for deals that limit competition, but hiring and licensing deals like this often slip past scrutiny.
Image by Firmbee.com, from Unsplash
2.3 Million Users Infected By Verified Chrome and Edge Extensions
- Written by Kiara Fabbri Former Tech News Writer
- Fact-Checked by Sarah Frazier Former Content Manager
A major security breach of browsers exposed more than 2.3 million users to malware, it did this through verified Chrome and Edge extensions that appeared safe.
In a rush? Here are the quick facts:
- Malware was hidden in verified and featured extensions with legitimate functions.
- Malware silently installed through updates after years of clean operation.
- Extensions hijacked browsers, tracked activity, and redirected users to fake sites.
According to research from Koi Security , 18 extensions in a campaign dubbed RedDirection secretly hijacked browsers, tracked user activities, and enabled additional attacks through trusted interfaces.
The main extension responsible for the security breach was “Color Picker, Eyedropper — Geco colorpick.” The extension delivered its promised functionality by providing a complete color selection feature. The extension operated as a color selection tool, but it secretly tracked all websites users visited, transmitted URL data to command and control servers, and redirected users to fake websites.
“This isn’t some obvious scam extension thrown together in a weekend” the researchers wrote.
“This is a carefully crafted trojan horse that delivers exactly what it promises (a functional color picker) while simultaneously hijacking your browser, tracking every website you visit, and maintaining a persistent command and control backdoor. Not only that, but it remained legitimate for years before becoming malicious through a version update,’’ the researchers noted.
Indeed, the researchers explain how these extensions were clean for years before malicious code was added through silent version updates, a move that took advantage of Google and Microsoft’s trust systems, including verification badges and featured placements.
“This isn’t just another malware discovery,” researchers said. “It’s proof that the current marketplace security model is fundamentally broken,” the research team added.
The RedDirection campaign included popular extensions that functioned as emoji keyboards, video speed controllers, VPN proxies, and dark themes, which appeared and operated like standard tools. The extensions operated as a single network through their identical malware structures and command servers, to steal login details, banking information, and install additional malware.
Koi Security advises users to eliminate untrusted extensions, while performing browser data cleaning, malware scanning, and account monitoring. The discovery raises doubts about Chrome and Edge’s extension verification process, and the ability of users to trust installed extensions.
“This is a supply chain disaster,” researchers warned.
