Cloudflare
Cloudflare business and news from across the web.
Latest coverage
'Every single week it's going to feel like the sky is falling': Cloudflare CEO says strap in for 'a couple years' of AI-powered internet security mayhem
Cloudflare CEO Matthew Prince predicts a challenging couple of years for internet security due to AI accelerating the discovery of vulnerabilities. He anticipates frequent security incidents similar to the recent PHP flaws, though believes this will ultimately lead to more secure software in the long term. Users are advised to stay updated on software patches and exercise caution.
PHP vulnerabilities lead to website shutdowns across the world ahead of security release
PHP is preparing to release an update to address two critical vulnerabilities, both rated CVSSv4 9.2. This announcement has prompted widespread website shutdowns, including government sites in Europe, as organizations prepare for potential cyber attacks. Cybersecurity agencies are urging users to update immediately or take their sites offline.
'We expect mass attacks to start within 15 minutes of the release': Web admins braced for emergency PHP updates to plug leaked high-severity holes
The PHP security team has announced an upcoming fix for two critical vulnerabilities, rated 9.2/10, with a fix expected at 2 PM UTC. Cybersecurity firms warn that attacks are anticipated to begin within 15 minutes of the release due to leaked information about the flaws. Companies like Cloudflare are already updating firewall rules to mitigate emerging threats.
Morning Safety Dance
Cloudflare has largely moved away from third-party security tools, a strategy not recommended for others. The article also highlights how artificial intelligence is being used as both a tool and a target in the latest wave of cyberattacks.
Sunday Metaverse
Cloudflare has revealed various causes for internet downtime, including hardware failures, software bugs, and external factors like power outages. The company shared insights into troubleshooting and maintaining stable internet connectivity.
Jensen Huang's first-ever post on X is in defense of open access to AI models, alongside Google, OpenAI, and Meta
Nvidia CEO Jensen Huang, alongside Google, OpenAI, and Meta, posted on X advocating against premature restrictions on open AI models. He argues that open models are crucial for AI's growth, innovation, and global accessibility, while acknowledging their inherent risks. The companies signed an appeal to avoid stifling competition and driving innovation overseas.
Evening Metaverse
Alphabet reported its Q2 2026 earnings, with YouTube's ad revenue showing a significant 13% growth. The article also touches on the end of a long period of fee listing requirements for Internet Service Providers and mentions Cloudflare's internal DNS.
Evening Metaverse
Cloudflare announced it will begin filtering web crawlers that serve AI companies. This move aims to prevent AI firms from scraping data from websites without permission.
Cloudflare will filter out web crawlers that serve AI companies
Cloudflare is implementing a new policy to filter out web crawlers that serve AI companies. This move aims to give website owners more control over how their content is utilized by AI firms for training purposes.
Morning Safety Dance
Web browsers and Cloudflare are collaborating to enhance defenses against malicious bots. Separately, a WIRED article highlights the increasing difficulty in spotting World Cup-related scams.
Morning Metaverse
Cloudflare's CEO has expressed concern that bots now constitute the majority of internet traffic, a milestone that was not anticipated until the following year. This surge in automated traffic, particularly 'agentic' AI, is reshaping online interactions and security concerns.
Web infrastructure company Cloudflare says Claude Mythos reasoning 'looks like the work of a senior…
Cloudflare has analyzed Anthropic's Claude Mythos AI model, finding it to be a significant advancement in identifying and exploiting software vulnerabilities. While the model can chain low-severity bugs into severe exploits, Cloudflare suggests a more directed, multi-agent approach is more effective than a single agent reviewing large codebases. The company believes the focus should shift from rapid patching to architectural defenses that make exploitation harder, even when bugs exist.
AI Yi-Yi!
Cloudflare has announced significant job cuts, reducing its workforce by 20%, which amounts to approximately 1,100 employees. The company attributes these layoffs to the integration and impact of artificial intelligence technologies. Following the news, Cloudflare's stock experienced a notable decline of 19%.
Out of the Blue
The website has implemented new measures, including the integration of Cloudflare, to address and block problematic traffic that was hindering legitimate access. This effort aims to improve the overall performance and stability of the site.
'Increasingly the distinction between bots and humans is moot' says one of the biggest web-infrastructure…
Cloudflare predicts a shift towards an agent-neutral web where the distinction between bots and humans becomes increasingly blurred. The company proposes solutions like Privacy Pass to verify behavior without compromising identity, aiming to prevent websites from being overwhelmed by AI data ingestion and maintain a monetizable online ecosystem.
'The most boring protocol on the internet is also, quietly, one of the most abusable': Doom has been broken down into 1,964 DNS text records and can be run via a PowerShell script
The classic game Doom has been successfully run by splitting its code into 1,964 DNS text records and reassembling them using a PowerShell script. This innovative method, developed by Adam Rice, highlights the potential for abuse within the DNS protocol, which was not designed for file storage. The project demonstrates how even seemingly mundane internet protocols can be creatively exploited.
Cloudflare claims 'today’s threat landscape is more varied and chilling than ever' in latest report
Cloudflare's latest threat report indicates a more varied and dangerous cybersecurity landscape, with sophisticated nation-state actors, hyper-volumetric DDoS attacks, and the misuse of legitimate cloud tools like Google Calendar and Dropbox. The report highlights the increasing complexity of threats, exacerbated by advancements in AI, and advises users to remain vigilant against phishing and suspicious links.