Mic Drop: Anne Neuberger on AI: ‘We have to challenge ourselves to be first’

Summary

Podcast Summary: Click Here – "Mic Drop: Anne Neuberger on AI: ‘We Have to Challenge Ourselves to Be First’"

Release Date: February 28, 2025

---

Introduction

In the episode titled "Mic Drop: Anne Neuberger on AI: ‘We Have to Challenge Ourselves to Be First’", Recorded Future News’ podcast Click Here features an in-depth conversation between host Dina Temple-Raston and Anne Neuberger, the former Deputy National Security Advisor for Cyber and Emerging Technologies under President Biden. Recorded at the Munich Security Conference—a pivotal gathering akin to Davos but focused on global policy—this episode delves into the intricate world of artificial intelligence (AI), cybersecurity, and the strategic competition between global powers, particularly the United States and China.

---

Deepseek: China's Cost-Effective AI Breakthrough

The conversation opens with a discussion about China’s AI startup, Deepseek, which garnered attention by developing one of the world’s most powerful AI systems with significantly reduced computing power.

Neuberger emphasizes the importance of innovation under constraints:

“I think Deepseek showed a very powerful lesson, which is we've learned this in any number of technologies. The tech is most what's the best mousetrap, and what's the easiest and cheapest to deploy.”
[00:59]

Deepseek's Achievement:

• Utilized a fraction of the ~16,000 specialized chips typically employed by AI companies.
• Published their methodologies in a series of papers, offering transparency and opportunities for other researchers to adopt cost-reducing techniques.

Implications:

• US AI firms are likely to integrate Deepseek’s innovations, enhancing efficiency and reducing costs in AI development.

---

Export Controls and Domestic Innovation in China

Dina Temple-Raston raises a critical issue regarding the effectiveness of US export controls:

“Some experts are saying that by restricting China's access to advanced chips, we're only accelerating their ability to do some sort of innovation domestically. Do you think that's true?”
[03:14]

Neuberger's Insight:

“We know that the export controls on chips certainly have made it harder for China to make advancements. We also know that when there are constraints, constraints often lead to innovation.”
[03:39]

Key Points:

• Innovation Under Pressure: Constraints from export controls have driven China to develop more efficient AI models.
• Data as a Competitive Edge: High-value data, such as images from millions of Hikvision cameras worldwide, play a crucial role in training advanced AI systems.
• Global Deployment of Chinese Tech: The widespread use of Chinese technology facilitates data accumulation, bolstering China’s AI capabilities without similar privacy protections found in regions like Europe.

---

AI in Cybersecurity: Offense vs. Defense

The discussion shifts to the dual role of AI in cybersecurity:

Cyber Offense:

“AI definitely accelerates any number of those steps [in cyber attacks].”
[05:23]

Neuberger explains how AI enhances offensive cyber capabilities:

• Vulnerability Detection: AI speeds up identifying weaknesses in code and targets.
• Customized Attacks: Examples include tailored spear-phishing emails that are more convincing and harder to detect.

Cyber Defense:

“AI can make a bigger difference, in my view, on helping cyber defenders if we move out quickly.”
[05:23]

Advantages for Defenders:

• Pattern Recognition: AI can detect anomalies and patterns indicative of cyber threats.
• AI vs. AI: Leveraging AI to identify and counteract malicious AI-generated activities.

Practical Example: Neuberger cites missile defense systems:

“Patterns that AI can use to help find where a launch might be... Anything that's generated by AI from a pattern perspective can be used on the flip side.”
[06:19]

---

Strategic Competition: Democracy vs. Autocracy

Dina Temple-Raston probes the impact of AI on potential conflicts, such as a US-China confrontation over Taiwan:

“So if you think about the possibility of conflict between the US and China over Taiwan, Neuberger says AI could have a profound effect on that fight.”
[07:00]

Neuberger’s Perspective:

“Advanced AI models that are trained on different kinds of ships, different kind of military practices, military maneuvers can make a big difference in being able to detect and conduct military operations.”
[07:09]

Key Insights:

• Military Applications: AI enhances capabilities in surveillance, detection, and strategic planning.
• Autocratic Advantages: Autocracies may deploy AI more rapidly due to fewer privacy constraints, giving them a strategic edge.

---

Reflections on Cybersecurity and Policy Innovations

Transitioning from her role at the White House, Neuberger reflects on past cybersecurity challenges and the necessity for innovative policy approaches:

Building Trust:

“Technology moved far faster than policy and law, than transparency about how we were using powerful technologies in ways that really mattered to American citizens.”
[09:08]

Need for Collaboration:

“We in the US and democracies have to ensure that we deploy in a way that retains the trust of our citizens and our allies.”
[09:08]

Rethinking Cybersecurity:

“The traditional way of looking at cybersecurity, let's be very candid, wasn't working.”
[10:29]

Innovative Policy Approaches:

• Ransomware Initiatives: Collaborating globally to target the financial underpinnings of ransomware attacks, particularly through cryptocurrency regulation.
• Counter Ransomware Initiative: Engaged 71 countries in capacity building and blockchain analysis to combat cybercriminals effectively.

Strategic Takeaway:

“We need to be far more agile and quick in trying new policy and defensive approaches.”
[13:15]

Recommendation:

• Embrace adaptable and transparent strategies that allow for rapid response and continuous improvement in cybersecurity defenses.

---

Concluding Thoughts and Future Directions

Anne Neuberger underscores the importance of proactive and innovative approaches in the face of evolving AI and cybersecurity challenges. She advocates for:

• Agility in Policy Making: Avoiding the paralysis of over-deliberation to implement effective defenses swiftly.
• Continuous Assessment and Transparency: Regularly reviewing and communicating policy measures to maintain public and allied trust.
• Emphasizing Collaboration: Strengthening partnerships between the private sector, government entities, and international allies to bolster cybersecurity resilience.

Final Quote:

“Sometimes good enough with a preparation to always go back and assess and review and consider and doing it transparently to explain.”
[14:21]

---

Conclusion

This episode of Click Here provides a comprehensive exploration of the current AI landscape, highlighting the strategic imperatives for democracies to innovate and collaborate in cybersecurity. Anne Neuberger’s insights reveal the delicate balance between leveraging AI for national security and mitigating its potential misuse by adversarial forces. As global competition intensifies, the need for agile, transparent, and cooperative policy frameworks becomes ever more critical to safeguarding democratic values and ensuring technological leadership.

---

Notable Quotes with Timestamps:

• Anne Neuberger:

  • “I think Deepseek showed a very powerful lesson...” [00:59]
  • “We need to be far more agile and quick in trying new policy and defensive approaches.” [13:15]
  • “Sometimes good enough with a preparation to always go back and assess and review...” [14:21]

• Dina Temple-Raston:

  • “Some experts are saying that by restricting China's access to advanced chips, we're only accelerating their ability to do some sort of innovation domestically. Do you think that's true?” [03:14]
  • “So if you think about the possibility of conflict between the US and China over Taiwan...” [07:00]

---

This summary encapsulates the pivotal discussions from the Click Here podcast episode, providing listeners and non-listeners alike with a clear understanding of the critical intersections between AI, cybersecurity, and global strategic dynamics.

Transcript

[Dina Temple Rastan]

From Recorded Future News and prx, this is Click Here. The Munich Security Conference is kind of like Davos, but for policy wonks instead of the rich and famous. Welcome to this year's Munich Security Conference. It was founded in 1963, and it brings together heads of state, diplomats and business leaders from the world's leading democracies for three days of meetings and presentations. And that's where I caught up with Anne Neuberger, the former deputy National Security Advisor for cyber and emerging technologies. She was in the Biden White House, and we ducked into a hotel coffee shop just outside the conference's security perimeter to talk about all things artificial intelligence, including the latest AI entry from China, Deepseek.

[Anne Neuberger]

I think Deepseek showed a very powerful lesson, which is we've learned this in any number of technologies. The tech is most what's the best mousetrap, and what's the easiest and cheapest to deploy.

[Dina Temple Rastan]

From recorded Future News, this is Click Here's Mic Drop, a longer listen to one of our favorite interviews of the week. I'm Dina Temple Rastan, and today a conversation with Anne Neuberger. We talked about everything from Deepseek, the Chinese startup that appears to have developed powerful artificial intelligence on the cheap, to how AI might aid the competition for the future world order. Stay with us. Click Here is brought to you by Progressive Insurance. Do you ever find yourself playing the budgeting game? Well, with the name your price tool from Progressive, you can find options that fit your budget and potentially lower your bills. Try [e-mail] progressive casualty insurance company and affiliates price and coverage match limited by state law. Not available in all states. I'm Dena Templewost, and this is Click Here's Mic Drop. Back in January, US Financial markets took a dive after China's Deep Seq announced that it had built one of the world's most powerful AI systems using significantly less computing power. Where AI companies use about 16,000 specialized chips, deep SEQ uses a fraction of that amount. Without getting into the details of how they did it, suffice to say they experimented with a way to use chips and it paid off. And because they laid out what they did in a series of papers, other AI researchers will likely use some of their methods to reduce the overall cost of building AI.

[Anne Neuberger]

They innovated a great deal, practically and technically in the way they generated the outcomes and in the cost of that. And I think that proved and indeed US AI companies will be adopting some of those innovations.

[Dina Temple Rastan]

Some of China's innovations came despite years of export controls aimed at ensuring that China wouldn't have the most advanced chips to develop their own AI. Some experts are saying that by restricting China's access to advanced chips, we're only accelerating their ability to do some sort of innovation domestically. Do you think that's true?

[Anne Neuberger]

I think there's some nuance there in that. We know that the export controls on chips certainly have made it harder for China to make advancements. We also know that when there are constraints, constraints often lead to innovation. So we certainly know that as a result, China has had to be more creative with regard to how do they build more efficient models, how do they generate the outcomes they're looking for with the computing constraint. And frankly, I think we really need to focus also on the role of data, specifically high value data. High value data training sets that lead to significant advancements.

[Dina Temple Rastan]

One of the key things here, she says, is that countries around the world use Chinese technology, and China vacuums up that data to use back home. They don't have the same privacy protections that say, Europe has.

[Anne Neuberger]

What makes China so capable in AI goes beyond the compute. It goes to also what's high value data. In that case, millions of hikvision cameras around the world deployed, generating images that can be used to train for facial recognition models that can be used domestically in China or because we know Chinese tech is deployed around the world as part of China's approach to strategic competition in tech. So in the race and competition between autocracies and democracies, AI will help the autocracies because their lack of stabilities and privacy controls means they can deploy faster.

[Dina Temple Rastan]

How concerned are you that AI models, particularly those developed outside the US could be used to conduct sophisticated cyber attacks against things like American infrastructure.

[Anne Neuberger]

So certainly, as you think about the steps needed to design and conduct a cyber attack, finding vulnerabilities in code, identifying a target, seeing where that target is vulnerable, and then conducting the attack, whether that's a country, whether that's a criminal for financial gain, AI definitely accelerates any number of those steps. And we're seeing various adversaries do things like build a customized Spearfish email. So certainly AI makes a big difference on offense. I personally think given the delta between offense and defense, it's so much easier to attack than to defend. AI can make a bigger difference, in my view, on helping cyber defenders if we move out quickly. In using AI for the cybersecurity mission.

[Dina Temple Rastan]

It'S particularly helpful to defend against cyber attacks generated by AI, because, perhaps ironically, one thing AI is very good at is sniffing out other AI.

[Anne Neuberger]

So AI can be used to help find patterns that can help tell you that something was generated by AI and that could apply to text, that could apply to images. I'll give a practical example. So when you think about, you know, missile launches and the need to detect a missile launch quickly in order to alert missile defense, one of the hardest parts of that is missile launches are often mobile. They're moved around to be stored and hidden. There are patterns that AI can use to help find where a launch might be. Who are the individuals who work on this mission to help missile defense systems? Anything that's generated by AI from a pattern perspective can be used on the flip side.

[Dina Temple Rastan]

So if you think about the possibility of conflict between the US and China over Taiwan, Neuberger says AI could have a profound effect on that fight.

[Anne Neuberger]

Clearly, advanced AI models that are trained on different kinds of ships, different kind of military practices, military maneuvers can make a big difference in being able to detect and conduct military operations. So that's a good example of where we can expect to see China applying AI as part of its strategic military planning.

[Dina Temple Rastan]

When we come back, Anne Neuberger talks about AI and the strategic competition between authoritarian regimes and democratic ones and how we may need to adjust our thinking not just about newfangled technology like AI, but more traditional national security problems, too. We'll be right back.

[Joseph Cox]

Hey, I'm Joseph Cox, the host of another podcast I think you'll like, the 404 Media Podcast. We're an independent news outlet covering the bleeding edge of technology. And every week we discuss our latest stories, whether that's how AI images are taking over Facebook and fooling people, how drugs are being sold on Instagram, or the spread of AI enabled surveillance cameras. We hold tech to account. Catch me and the rest of the 404 Media gang wherever you listen. Just search for the 404 Media podcast. Chat to you soon.

[Dina Temple Rastan]

So back in the summer of 2013, this thing happened that rocked the intelligence community.

[Edward Snowden]

My name's ed Snowden.

[Joseph Cox]

I'm 29 years old.

[Dina Temple Rastan]

He's the former National Security Agency contractor whose leak of agency documents set off a national debate over the proper limits of government surveillance. Anne Neuberger was actually at the NSA assessing risks against the agency shortly after that happened.

[Anne Neuberger]

And when I look back and I saw firsthand the loss of trust by the American people and divisiveness between government and the US tech sector. That came because technology moved far faster than policy and law, than transparency about how we were using powerful technologies in ways that really mattered to American citizens. In Ways that mattered to partners around the world. So I walked away believing that that was avoidable and that we had to find a way to work together. And I think given that in AI, the engine, the partnership really is the private sector, this is upon us as a country to really consider that from the outset in the way we deploy AI in the most sensitive domains. Because I think it is really critical that given we will see AI, AI is an advancement, we can expect to see it a part of countries, companies, strategies. I think we in the US and democracies have to ensure that we deploy in a way that retains the trust of our citizens and our allies. As a democracy, you know, I often think about we are in a competition for the future of the world order, and a part of that is who we are and our values as a democracy, as well as the power of our country as a home for the most creative work.

[Dina Temple Rastan]

And she said, in some ways, when it comes to cybersecurity, we haven't been creative enough. And she'd like to see that change.

[Anne Neuberger]

The traditional way of looking at cybersecurity, let's be very candid, wasn't working. And I think as we look at that and we say, what are the creative approaches we need to do? How do we approach this with the key partners, which are both private sector, both the entities that need to be protected, hospitals, the companies that produce the tech, and then other countries, and how do we do it in the context of who are the actors and how do we make it riskier, costlier, and harder for them?

[Dina Temple Rastan]

This is something she thought a lot about in her job at the White House as Deputy National Security Advisor. Take ransomware, for example.

[Anne Neuberger]

The set of policy approaches ranged from helping to make hospitals harder targets, helping to ensure that companies who build technology build more secure tech from the outset so it's easier for hospitals to deploy that. And then saying, how do you take down adversaries infrastructure? How do you freeze and seize cryptocurrency, which is often used to make those payments. In fact, we had to work with governments around the world because we were all dealing with this.

[Dina Temple Rastan]

And they eventually began to zoom in on the financial component of the problem, something the US had success with in the past with things like terrorism financing. Back in the early 2000s, they had a lot of trouble trying to stop terrorists from using the informal hawala network to finance their attacks.

[Anne Neuberger]

Countering terror financing was a space that the US worked hard, doubled down, and then really had an impact. Cryptocurrency has become the new hawala it funds all kinds of transnational crime. It is the way that countries launder money, whether it's terror financing, whether it's weapons profiteering, et cetera, focusing and doubling down and saying, okay, how do we work with virtual asset service providers to ensure that at the very least funds can be frozen, Share bad wallet information, work with foreign governments around the world to help train them in understanding how to do blockchain analysis? All of that work was underway. A big part of it happened under the Counter Ransomware initiative, which brought together 71 countries to do capacity building and blockchain analysis to put in place anti money laundering in cryptocurrency, all in parallel to doing that strategic work. Practical efforts to put pressure on the really small set of actors doing this work needed to move far faster.

[Dina Temple Rastan]

So you've had a little time away, some time to reflect on all these things you've been working on for so many years. What's your big takeaway now?

[Anne Neuberger]

So as we think about the larger strategic competition for the future of world order, the competition between the US and China, Russia, an increasingly troubling bad actor in a number of hybrid ways, I think sometimes we're far more prepared to use advanced technology for our national security missions versus to defend against adversary use of it. We need to be far more agile and quick in trying new policy and defensive approaches. And that's really reflecting on that, a call to action for us to be willing to try things, take a pause, explain why, take a pause, see what we learn and move forward versus sometimes carefully deliberating on all of the puts and takes to where it takes us far longer to protect ourselves from an either country or criminal use of advanced technology that impacts companies, that impacts government's operations in a way that's really we shouldn't be tolerating.

[Dina Temple Rastan]

Sort of seems like you're saying we're letting the perfect be the enemy of the good.

[Anne Neuberger]

That's a good summary sometimes. And reflecting on it sometimes good enough with a preparation to always go back and assess and review and consider and doing it transparently to explain.

[Dina Temple Rastan]

From recorded future news this has been Click Here's Mic Drop. It was written and produced by Erika Gaeda, Megan Dietre, Sean Powers and me, Dina Temple. Rest it was edited by Karen Duffin. We'll be back on Tuesday with an all new episode of Click Here. Have a great weekend.

[Edward Snowden]

If you're looking for a daily guide to cybersecurity news and policy, sign up for the Cyber Daily. From Recorded Future News. It serves up today's most interesting and important cyber stories from our sister publication the Record, and then aggregates all of the big cyber stories you might have missed from news outlets around the world. Just go to the Record Media and click on Cyber Daily to get all you need to know about the world of cybersecurity right in your inbox.