Angelo Dalli: How One Software Update Grounded the World and How AI Can Save Us
Yesterday, the world experienced a digital meltdown when a faulty update from cybersecurity firm CrowdStrike caused chaos on a global scale. This single update led to grounded flights, disrupted health services, and crashed payment systems. Here’s what happened and how AI can help prevent such tech disasters in the future.
The Incident
CrowdStrike, a major player in cybersecurity, pushed an update to their Falcon software that went horribly wrong. The update triggered what’s known as the dreaded “Blue Screen of Death” on Windows computers. Millions of machines were affected, leading to widespread disruptions across various sectors.
CrowdStrike’s Falcon software is designed to protect computers against cyber threats. In order to do so, it needs to have permission to have deep system access on protected devices. The faulty update involved something called a kernel driver, which interacts with the very core of the operating system – think of it as the heart of the operating system.
This update caused Windows computers to enter a catastrophic spiral of reboots, triggering a system crash on affected computers, and eventually leading to the global outage.
Unfortunately, the update was pushed simultaneously to millions of machines, leading them all to crash at the same time.
View this post on Instagram
My Experience
As an AI expert, I found myself caught in the middle of this chaos. I was stuck at Malta International Airport for six hours because of the tech glitch.
Witnessing the confusion and frustration firsthand made it clear just how critical reliable software updates are for our interconnected world. This incident highlights our heavy dependence on a few tech giants like Amazon AWS, Microsoft Azure, and Google Cloud. When a single point of failure occurs, the consequences can be massive. The CrowdStrike fiasco is a stark reminder of the vulnerabilities in our digital infrastructure.
The Role of AI
Artificial Intelligence offers a promising solution to prevent such incidents. AI can revolutionize the way we test and deploy software updates.
Automated testing can run thousands of test cases, catching potential issues before they become problems. New forms of AI like Explainable AI and Neurosymbolic AI can help prevent future recurrences.
Explainable AI helps developers understand the AI’s decision-making process, making it easier to debug and fix issues. Neurosymbolic AI, which combines neural networks with logical reasoning, enhances testing capabilities and error detection. Neurosymbolic AI can generate and verify test cases based on logical rules and learned patterns, improving the thoroughness and accuracy of testing.
These systems can trace the logic behind failures, offering precise insights into why an update might cause problems.
View this post on Instagram
Preventative Measures
To avoid future tech disasters, phased rollouts of updates should be implemented. This approach allows for early detection of issues in a controlled environment.
Additionally, simple rules such as avoiding updates near weekends ensure that technical support teams are available to address any problems that arise. AI-powered monitoring can predict potential failures and automatically trigger responses to mitigate issues, such as rolling back updates or reallocating resources.
AI can also significantly enhance the software testing process by automating repetitive and complex testing tasks. By analyzing existing code and test cases, AI tools can generate new test scenarios, ensuring that all aspects of the software are thoroughly tested.
Conclusion
The CrowdStrike incident shows us the importance of leveraging AI to make software updates safer and more reliable. By adopting these advanced AI solutions and best practices, we can build a more resilient digital infrastructure and prevent similar disruptions in the future.
If these problems are left unresolved, the consequences could be dire. Future outages could become more frequent and severe, potentially affecting critical infrastructure like hospitals, financial systems, and government services. This could lead to widespread chaos, financial losses, and even threats to public safety. Our reliance on technology makes it imperative to address these vulnerabilities now.
The future of our digital world depends on our ability to adapt and implement these advanced solutions to safeguard against such catastrophic failures.
Angelo Dalli is UMNAI’s Chief Scientist and Co-Founder. As an Artificial Intelligence
expert and serial tech entrepreneur, he combines deep technical ability with commercial
acumen. Angelo has founded nine startups, was involved in four IPOs/M&As and invested in
over forty successful companies in the technology, transportation, entertainment and fintech
industries. Angelo has created foundational neuro-symbolic AI technology, called Hybrid
Intelligence, that combines human-like reasoning with hypergraph neural networks for
smarter and trustworthy AI solutions. He has authored over 20 peer reviewed publications,
published 52 granted/pending AI patents, and is a member of EBAN, YPO, and a WBAF
senator. Guided by the philosophy of creating scalable and innovative technology businesses
with a significant social impact, Angelo’s contributions have not only advanced the field of
AI but have also fostered positive change.
Lovin Malta is open to interesting, compelling guest posts from third parties. These opinion pieces do not necessarily reflect the views of the company. Submit your piece at [email protected]