The AI juggernaut hits cybersecurity, open source and IT infrastructure
The week in enterprise tech brought us insights from three major conferences: RSA Conference in San Francisco, Red Hat Summit in Denver, and Boomi World in the same city.
Developers and executives gave us a glimpse into the state of cybersecurity, open source, and IT infrastructure. And AI was, unsurprisingly, top of mind at each event.
Here's a breakdown of the key themes from the week:
- The defense is winning the AI-powered cyberwar
Generative AI is being used to create phishing emails, but it's also aiding the defenders. The speed at which AI-powered tools allow defenders to respond to threats outweighs the advantage gained by the attackers.
Erik Bradley, VP of research at Enterprise Technology Research, noted that fewer than 10% of enterprises are reducing the number of security providers they use despite a steady stream of M&A.
Why? Innovation is happening faster than consolidation.
Chris Krebs, chief intelligence and public policy officer at SentinelOne, said: "The defense is winning right now." And Kevin Mandia, CEO of Mandiant at Google Cloud, agreed: "We're responding much faster to attacks."
However, Andrew Weiss, chief technology officer at Weiss Analytics, warned: "There is a ton of risk."
And CEO of SentinelOne, Tomer Weingarten, added: "The offense is going to have a longer tail to gear up."
- Securing AI will be a big deal this year
AI security is set to become a significant issue in the wake of the increasing number of people interacting with large language models.
Professor Adi Shamir, the Borman professor of computer science at the Weizmann Institute, said: "The new thing this year is securing the AI."
And cybersecurity is crucial to securing AI, according to Chris Doman, security researcher at Check Point Software. He said: "If you think about the way AI is often trained, it's on the basis of ingesting as much data as possible. A lot of that data may not be very sensitive or important, but some of it certainly will be. As more data is ingested and used to train models, the potential for sensitive data being included in that increases."
And as the models become more advanced, so do the attacks.
- The AI effect There was no let-up in AI news this week, though the focus was more on business fundamentals than the tech itself.
Mistral AI is reportedly seeking to raise $600m at a $6bn valuation.
Microsoft and Amazon Web Services are building data centers for AI in Wisconsin and Singapore, respectively, with investments of billions.
And Arm's earnings guidance raised fears among investors that some AI stocks may be overpriced.
- The week in cyber
The RSA Conference was an opportunity to take the pulse of the cybersecurity industry. And, as ever, there was a lot going on.
Despite headlines about the escalating frequency and scope of cyberattacks, there was some good news. Generative AI is helping the defenders more than the attackers, in part by allowing defenders to act more quickly to counter the skyrocketing number of novel vulnerabilities.
Chris Krebs, chief intelligence and public policy officer at SentinelOne, said: "The defense is winning right now." And Kevin Mandia, CEO of Mandiant at Google Cloud, agreed: "We're responding much faster to attacks."
But it's no time for complacency, obviously, especially as we