Today on Product Saturday: Cloudflare introduces a new set of security tools for AI applications, HPE tackles data lakes in hybrid clouds, and the quote of the week.
Today: Writer CEO May Habib explains her company's generative AI strategy, Ampere and Arm get a little cozier, and the latest enterprise moves.
Writer has raised $326 million to build a "full stack" approach to generative AI by developing its own LLMs and building application tools around those models. CEO May Habib thinks using synthetic data and a collaborative process will spur enterprise AI adoption.
Today on Product Saturday: Cloudflare introduces a new set of security tools for AI applications, HPE tackles data lakes in hybrid clouds, and the quote of the week.
Welcome to Runtime! Today on Product Saturday: Cloudflare introduces a new set of security tools for AI applications, HPE tackles data lakes in hybrid clouds, and the quote of the week.
(Was this email forwarded to you? Sign up here to get Runtime each week.)
Lock it down: The rush to build generative AI applications seems very likely to lead to security problems down the road, given the pace of development and the unfamiliarity that most software organizations have with a new and different way of building apps. Cloudflare does many things these days, but it started off as a cloud security company, and this week unveiled a new set of security tools designed specifically to protect AI applications.
With Cloudflare for AI, "Cloudflare customers will be able to protect themselves against the most pressing threats facing today’s AI models, including employee misuse of tools, toxic prompts, personally identifiable information (PII) leakage, and other emerging vulnerabilities," the company said in a press release. The new tools help companies discover where AI apps are running on their networks and monitor their performance.
The watchmen: As AI applications mature and the cost of inference comes down, companies concerned about privacy and security that have already invested in their own data centers will want to run those apps internally. LogicMonitor introduced several new features for its observability service this week that will help those companies track the performance of those apps as well as how much they cost to run.
The company added better support for Nvidia's GPUs as well as Kubernetes, which will make it easier for companies pursuing hybrid cloud strategies to look across their entire operating environment. LogicMonitor also updated its AI agent, which helps users filter and prioritize alerts coming from the platform.
Data, data everywhere: This edition of Product Saturday could have been devoted entirely to the dozens of announcements made at GTC this week by Nvidia and its partners, but that's no fun. One announcement that did fly under the radar, however, was the Nvidia AI Data Platform, which is "a customizable reference design that leading providers are using to build a new class of AI infrastructure for demanding AI inference workloads: enterprise storage platforms with AI query agents fueled by NVIDIA accelerated computing, networking and software," the company said in a press release.
Companies that want to deploy generative AI applications have quickly realized that the first thing they have to do is get their corporate data in order, and the ones that still manage their own data centers are probably long overdue for an update to their data storage strategy. As it so happens, dozens of storage vendors lined up to support Nvidia's new reference design.
Connect the dots: An important part of getting corporate data ready for AI involves bringing it all together in a centralized location, which is easier said than done given how many applications and services the average company is using. Companies use Airbyte to build data pipelines that help move everything along, and this week it introduced several new features as part of its Winter 2025 update.
"This release is intended to provide critical features required for you to use your first party data in AI tools and turn them into your competitive advantage," the company said in a press release. The update includes connectors to data platforms like Oracle and SAP, support for AWS PrivateLink, and support for Iceberg, the open table format that the data industry is quickly making into a standard.
Makin copies: Ransomware attacks have faded somewhat from the public eye over the last year — there's a lot going on right now — but they remain an existential threat to companies that lack proper backup and recovery tools. Eon introduced a new cloud service this week that could help victims of ransomware get back in business.
"By enabling smart restores of clean data in minutes, as opposed to hours or even days, Eon delivers unprecedented speed and security for modern cloud environments, effectively reducing the impact of ransomware by minimizing downtime and ensuring business continuity," the company said in a press release. Eon, founded by several ex-AWS engineers, has raised $127 million in funding.
They're called the Big Three for a reason: New data from Synergy Research shows that at the end of 2024, AWS, Microsoft, and Google Cloud accounted for "59% of all hyperscale data center capacity" in the world, and 54% of the world's data center capacity is located in the U.S.
"AGI doesn't mean God. It means super, super good." — Writer co-founder and CEO May Habib, weighing in on the quixotic quest that some AI devotees believe will arrive shortly.
Meta gets paid when people use its Llama LLMs through one of its partners, which include The Big Three, Databricks, and Snowflake, according to a court filing.
Microsoft narrowly avoided an eight-week losing streak for its stock price, which would have been the first time that had happened since 2008, according to CNBC.
Thanks for reading — see you Tuesday!
Today: Writer CEO May Habib explains her company's generative AI strategy, Ampere and Arm get a little cozier, and the latest enterprise moves.
Today: Why Google just made Wiz its largest acquisition ever, Nvidia unveils the next generation of its GPU lineup, and the latest funding rounds in enterprise tech.
Today on Product Saturday: A popular programming language just got a big boost from Microsoft, Crusoe adds new services to its AI cloud, and the quote of the week.
Today: why companies building generative AI applications might not need to wait for perfection, The Linux Foundation gets bigger, and the latest enterprise moves.
