Satya Nadella warns of unfixable vulnerabilities as AI-driven cyberattacks accelerate

Exercise 1
Vocabulary
Repeat each word, its meaning and the example sentence out loud.
unpatchable
adjective/ʌnˈpætʃəbəl/unable to be fixed or updated with security software; permanently vulnerable without replacing the system entirely
The team discovered that the legacy HVAC controllers were unpatchable, forcing us to explore network isolation as a mitigation strategy.
firmware
noun/ˈfɜːrmwer/low-level software embedded directly in hardware devices that is difficult or impossible to update remotely
Our security audit found that the firmware on half our data center infrastructure had not been updated in over five years.
latent bug
noun phrase/ˈleɪtənt bʌɡ/a hidden defect in code that exists but has not yet caused problems, often triggered by specific conditions
The incident response team traced the outage to a latent bug that only manifested under peak load conditions.
threat vector
noun phrase/θret ˈvektər/a specific pathway or method that attackers could potentially exploit to compromise a system or network
We need to reassess our threat vectors now that attackers are using AI to automate reconnaissance across our supply chain.
in the wild
phrasal expression/ɪn ðə waɪld/actively being exploited or encountered in the real world by actual attackers, not just theoretically possible
Once a zero-day exploit circulates in the wild, we typically have only hours to deploy patches across our infrastructure.
Exercise 2
Article
Read the article aloud on your own, or listen to it first.
Satya Nadella warns of unfixable vulnerabilities as AI-driven cyberattacks accelerate
Tap any word to see what it means.
Satya Nadella, chief executive of Microsoft, delivered an unsettling assessment of global infrastructure security at the New York Times' DealBook Summit on Wednesday.
Nadella emphasized that vast portions of critical infrastructure contain firmware and embedded systems that simply cannot be updated or repaired, yet these same systems remain connected to the internet.
He cited cooling equipment in data centers, HVAC systems, and transportation infrastructure as prominent examples of devices running unfixable software that nonetheless face exposure to networers.
The root cause, according to Nadella, stems from what he termed the "smartification of the economy"—the phenomenon where traditionally disconnected devices were retrofitted with internet connectivity without adequate security consideration.
The discussion came amid a series of high-profile global outages affecting major technology firms and government agencies, though officials have disputed whether these stemmed from coordinated cyberattacks.
Nadella touched on a recent Outlook incident, attributing it to a latent bug that created a "perfect storm" of operational failures, and committing to extracting architectural and operational lessons from the incident.
The fundamental dilemma, as Nadella articulated it, revolves around the arms race between patch velocity and exploit sophistication.
Historically, development teams could take time to thoroughly test patches, comprehend root causes, and orchestrate careful rollout procedures across their estate.
Today, artificial intelligence has democratized the creation of sophisticated cyberattacks, compressing the window for deliberate patch deployment into an impossible timeline.
Patches must now transition from availability to production deployment almost instantaneously when exploits circulate in the wild, without the luxury of comprehensive testing.
This forced acceleration, Nadella cautioned, inevitably introduces instability into systems that organizations depend upon for critical operations.
Nadella concluded by suggesting that the path forward requires both modernizing vulnerable architectural foundations and developing the capability to detect and defend threats in real time, rather than relying solely on remediation after the fact.
Discuss this story out loud with our AI speaking coach.
It asks one question at a time, at your level, and gives you feedback at the end.
Start the discussion Retell it3,000+ learners
Also works inChatGPTandClaude
Exercise 3
Discussion
Have a discussion based on the following questions.
- 1.In your own organization, how many legacy systems run firmware or embedded software that likely cannot be patched, and how does your company currently manage the security risks they pose?
- 2.Given that AI-accelerated exploits now outpace traditional testing cycles, what changes would need to happen in your development and deployment practices to operate more safely?
- 3.Nadella distinguishes between patching vulnerabilities after the fact and developing real-time detection capabilities. Which of these approaches does your organization prioritize, and what trade-offs have you observed?
- 4.Consider the devices in your workplace that are connected to the internet but designed when internet connectivity was not a primary concern. What security conversations have your teams had about their vulnerability?
- 5.If perfect patching is now impossible, what organizational or strategic shifts might companies need to make beyond purely technical solutions?
Exercise 4
Further discussion
Have a discussion based on the following questions.
- 1.How should responsibility for security be distributed when vulnerabilities stem from architectural decisions made decades ago before the internet pervaded every device? Who bears the cost of modernization?
- 2.To what extent should companies prioritize building resilience and fault tolerance into systems as an alternative to assuming they can remain completely secure from attack?
- 3.If critical infrastructure contains unpatchable components, what regulatory frameworks or industry standards should be established to manage this inherent risk?
- 4.How can organizations develop the specialized expertise required to detect and defend against threats in real time, rather than relying primarily on vulnerability disclosure and patch deployment cycles?
Discuss this story out loud with our AI speaking coach.
It asks one question at a time, at your level, and gives you feedback at the end.
Start the discussion Retell it3,000+ learners
Also works inChatGPTandClaude
Source
This lesson is our own retelling, for English learners, of a story from TechCrunch.
Related stories

Signal stops users taking screenshots to protect privacy from AI
22 November 2026 · B1

Signal Blocks AI Assistants from Reading Your Messages
21 November 2026 · B2

Major US Systems Go Down: Cyber Attack or Bad Luck?
21 November 2026 · B1

Researcher Exposes Massive Data Leaks Across AI Supply Chain Using Digital Tripwires
21 November 2026 · C1



