AMD says it’s going to invest up to $5 billion in Anthropic, while helping to expand the AI company’s computing power, according to an announcement on Wednesday. As part of the new partnership, Anthropic will deploy up to 2 gigawatts of AMD’s Instinct MI450 AI GPUs using the chipmaker’s new Helios rack-scale system, as reported earlier by The Wall Street Journal.

The companies plan to deploy the first gigawatt in the first half of 2027, building upon recent data center deals Anthropic has reached with SpaceX and TeraWulf. Anthropic has also signed AI infrastructure deals with Google, Broadcom, and Amazon, with rumors suggesting that it could reach an agreement with Meta as well.

AMD and Anthropic will also launch a “multi-year engineering collaboration,” where AMD will use Anthropic’s Claude across its software development, engineering, and product development. “By partnering with AMD across the stack, we are securing the capacity we need and optimizing it for training and serving Claude,” Tom Brown, Anthropic cofounder and chief compute officer, says in the press release.

OpenAI says its AI models mistakenly breached open-source AI platform Hugging Face during internal testing. In a blog post on Tuesday, OpenAI writes that GPT-5.6 Sol and “an even more capable pre-release model” discovered vulnerabilities within their sandboxed testing environment, allowing them to gain access to the internet and target Hugging Face.

On July 16th, Hugging Face disclosed a security incident that it says was driven by “an autonomous AI agent system.” Hugging Face’s AI agents detected and stopped the breach, which OpenAI has now admitted occurred during an evaluation of its models’ cybersecurity capabilities. OpenAI says “all evidence suggests that the models were hyperfocused on finding a solution for ExploitGym,” a benchmark system that measures whether AI models can turn security vulnerabilities into exploits.

As part of efforts to complete the evaluation, the AI models gained access to the internet by exploiting a zero-day vulnerability in the sandboxed environment. From there, OpenAI says its models “inferred that Hugging Face potentially hosted models, datasets and solutions for ExploitGym,” and then “searched for and successfully found ways to gain access to secret information that it could use to cheat the evaluation:”

In one example, the model chained together multiple attack vectors, including using stolen credentials and zero-day vulnerabilities to find a remote code execution path on the Hugging Face servers.

But as serious as this incident is, OpenAI appears to be using the “unprecedented” attack as an opportunity to make its AI systems look good — especially as it competes with cybersecurity rivals, like Anthropic’s Mythos and Gemini Flash 3.5 Cyber. OpenAI’s blog post has a chart showing how GPT-5.6 Sol is getting better at sustaining multistep cyber operations and also encourages enterprise customers to sign up to access its “Cyber” security model.

OpenAI adds that it’s now working with Hugging Face to investigate the security incident and will implement new controls within its research environment.

Google is launching Gemini 3.6 Flash alongside a new security model dedicated to quickly finding and patching security vulnerabilities. In a blog post on Tuesday, Google describes Gemini 3.5 Flash Cyber as a “cost-efficient and highly capable alternative” to larger, more expensive AI systems, such as the one offered by Anthropic’s Mythos.

The cybersecurity model is built upon Gemini 3.5 Flash and will be available first to governments and trusted partners via CodeMender, Google’s security-focused coding agent. As noted by Google, CodeMender can call upon 3.5 Flash Cyber “multiple times at high speed and low cost,” allowing the AI agents to scan more code paths and find vulnerabilities.

Anthropic’s Mythos 5 is a powerful AI security model released as part of the company’s Project Glasswing initiative. The compute-heavy model is expensive to use, costing twice as much as Claude Opus 4.8. Microsoft, which adopted Mythos for its security checks, had its biggest Patch Tuesday this month after using AI to find vulnerabilities. Other companies, like Google, are racing to keep up, while China’s Z.ai claims its model can compete with Mythos.

In addition to 3.5 Flash Cyber, Google says its upgraded Gemini 3.6 Flash offers improvements in coding and multimodal performance, while its new 3.5 Flash-Lite serves as the “most cost-effective” model in the 3.5 series.

Meanwhile, Google says 3.5 Flash Cyber achieved “competitive performance” compared to “significantly larger models” on the CyberGym AI cybersecurity benchmark when called upon up to five times. It also identified 55 “unique confirmed issues” in the V8 JavaScript Engine, compared to 47 found by Gemini 3.5 Flash and 36 by Opus 4.6. Google adds that 3.5 Flash Cyber found 10 issues that no other model discovered, noting that the model continued to find new code paths and vulnerabilities after being invoked multiple times.

Update, July 21st: Added information about Google’s other Gemini models.

Netflix says roughly 300 titles on its platform used generative AI, most of which occurred in post-production. The streaming service revealed the news in its second-quarter earnings report released on Thursday, saying it’s “increasingly leveraging these tools to deliver higher quality output more quickly and at a lower cost.”

It also provided some examples of titles that used AI, including The American Experiment, Glory, and Brasil 70: A Saga do Tri. These shows used the technology to “create highly complex sequences,” including “enhanced crowds, historical battle sequences, and worldbuilding establishing shots.” During Netflix’s call with investors, co-CEO Ted Sarandos said The American Experiment docuseries includes 17 minutes of “AI-enhanced footage,” which “were produced twice as fast and at half the cost of previous options.”

“In many of the cases, productions would have left out those key shots because they just wouldn’t have been able to afford them,” Sarandos added. “They wouldn’t have been able to do them in the timeframes that they’re working on.” Sarandos similarly said last year that AI was used to create a scene in the sci-fi series The Eternaut to save time and cut costs.

The streaming giant has begun to invest more heavily in AI as the technology becomes more advanced, with Netflix acquiring Ben Affleck’s AI startup and creating an AI animation studio. The service is also using the AI-generated voice of Gene Wilder in its new Wonka’s The Golden Ticket reality show.

Netflix reported earning $12.56 billion over the past few months, and says it’s still on track to double its ad revenue to $3 billion. In its letter to shareholders, Netflix also addressed some concerns about engagement, which came up after a report from Bloomberg revealed that the streaming giant is struggling to keep viewers around for the second season of its shows.

The service says “time spent is just one aspect of strong engagement,” adding that “quality and variety also matter.” It also highlights that its latest What We Watched report shows that subscribers watched over 97 billion hours, up 2 percent year over year. The company also announced that it will now switch to publishing this report just once per year, instead of twice.

Netflix has started to introduce new types of content in a bid to compete with free-to-watch services like YouTube. In the past year, Netflix has rolled out video podcasts, TikTok-style clips, and most recently announced plans to stream videos created by digital media brands, like BuzzFeed, which would typically appear on YouTube. Earlier this month, The Wall Street Journal reported that Netflix is considering adding always-on channels.

Update, July 16th: Added information from Netflix’s earnings call.