Posts tagged with “big tech”

Gemini 4 Argon: Google zet in op defensieve cyber-AI

Google heeft een nieuw vlaggenschipmodel, en dat komt niet zomaar voor iedereen beschikbaar. Alphabet, het moederbedrijf van Google, heeft Gemini 4 Argon gepresenteerd, een model voor algemeen gebruik dat programmeren, onderzoek en schrijven bestrijkt. Toch legt Google de nadruk vooral op een terrein: cyberbeveiliging.

De toegang is voorlopig beperkt. Argon gaat alleen naar een kleine groep cyberpartners van Google via het Fairwind Program, het beveiligingsinitiatief van het bedrijf. Volgens Google is het model speciaal getraind voor defensief cyberwerk en kan het "autonomously find, validate, and patch critical software vulnerabilities" - oftewel zelfstandig kritieke kwetsbaarheden in software opsporen, valideren en dichten.

Wat Argon volgens Google kan

Read More


Gemini 4 Argon: Google Bets on Defensive Cyber AI

Google has a new flagship model, and it is not being released to everyone. Alphabet, Google's parent company, has introduced Gemini 4 Argon, a general-purpose model that covers coding, research and writing. Google is putting most of its emphasis on one area, though: cybersecurity.

Access is restricted for now. Argon is going only to a small group of Google's cyber partners through the Fairwind Program, the company's security initiative. Google says it trained the model specifically for defensive cyber work and claims it can "autonomously find, validate, and patch critical software vulnerabilities."

What Google says Argon can do

The claims fall into three groups.

Read More


Meta's AI Data Center Tax Credit Saved $3.9B in 2025

Meta is cutting its US tax bill by billions of dollars with an unusual label. According to the New York Times, the company classifies its AI data centers as "pilot models" and the Nvidia chips inside them as experimental materials. That lets it claim a federal research tax credit meant to reward research and development.

The amounts are large, and they are rising. Meta saved $3.9 billion through the credit in 2025. That is up from $2 billion in 2024 and $700 million in 2023. The Times reports that no other publicly traded company benefits more from this credit.

Experiments at multi-gigawatt scale

Read More


Reddit Ends RSS Feeds and Public API, Citing AI Scrapers

Reddit is closing two of its most open doors. On Wednesday, the company said it will stop supporting RSS feeds on November 13 and shut down its public API by March 2027. The reason it gives is automated scraping, and the timing lines up with the growing value of Reddit's data to AI companies.

The news came as part of a larger set of updates for moderators and developers. For people who still read the web through feeds, it is an unwelcome change.

Why RSS is going away

Reddit describes RSS as a "common surface for large-scale scraping and automated abuse." In its view, the feeds have become an easy way for bots to pull content off the platform in bulk.

Read More


OpenAI Skips Nvidia's Open Agent Safety Platform: Why

Nvidia's new push to contain rogue AI agents launched on Monday with more than 100 companies on board. OpenAI was not one of them.

Other major players are also missing. Amazon, Google and Apple have not joined either. But OpenAI's absence stands out most, because its main rival, Anthropic, is listed as a supporter. Supporters are presumably expected to use and sell some version of the technology and to contribute features back to the project.

OpenAI has not rejected the effort outright. A company spokesperson told TechCrunch that OpenAI supports Nvidia's work. The two companies are already collaborating on agent security, including on one of the platform's core pieces of software.

Read More