There is a slow adoption of the shift to the left, endemic risks related to identity management, and an overly cautious deployment of AI in enterprises.
Sysdig, the leader in powered cloud security through real-time execution analysis, announces the conclusions of the Sysdig Report 2024. Drawing on real data, this seventh annual report details the dangerous practice of companies putting convenience ahead of preventive security in an effort to accelerate application development.
69% of companies have not yet integrated AI into their cloud environment. Only 31% of companies have integrated AI frameworks and modules, but only 15% of these integrations are used for generative AI tools like large language models (LLM). Considering the risk acceptance described in this report, organizations ignore security best practices but remain cautious when it comes to implementing of artificial intelligence.
91% of runtime checks fail. For left-shift safety, organizations perform early and frequent reviews during the development phase, recognizing build flaws, fixing the code, and then deploying again. The goal is to detect issues before delivery and before they become exploitable by attackers.ts. However, with 91% of runtime scans failing, teams seem to rely more on threat detection than prevention.
Only 2% of granted permissions are used. Identity management, for humans and machines, has become the most overlooked attack risk in the cloud. This is an opportunity for companies to improve their security posture, especially in light of known attacks in 2023 that leveragedi of permissive identities. In 2022, Sysdig found that 90% of permissions were unused, compared to 98% in 2023.
Reducing container lifetime doesn’t stop attackers. Attackers have near-instant knowledge of cloud environments, giving them the opportunity to move laterally. This is due to discovery automation and recognition, as well as the homogeneous nature of cloud environments. Running vulnerable workloads, no matter how fleeting, puts companies at risk of attack.
“Attackers are leveraging automation to exploit any vulnerabilities they can unmask,” said Crystal Morin, cybersecurity strategist at Sysdig. “The report shows that many companies seek faster innovation at the expense of more comprehensive security: a gamble that poses real risks to the survival of companies.”
“While I am not surprised by the apprehension surrounding the security of new technologies like AI, I am disheartened by the enormous number of excessive permissions being administered, especially for machine identities,” said Anna Belak, Director of the Office of Technology Strategy. Sysdig Cyber Security. “It’s a bit like obsessing about a plane crash when you regularly run red lights without a seatbelt. »
Based on analysis of millions of containers and thousands of cloud accounts, users, and roles, the report explores how companies of all sizes and industries around the world are using and securing cloud and containerized environments.