Interos

Forget the Super Bowl, the Real Monday Morning Quarterbacks are Targeting AI

Posted on January 29, 2025February 3, 2025 by Andrea Little Limbago

Author: Dr. Andrea Little Limbago, SVP, Applied AI, interos.ai

“DeepSeek R1 is AI’s Sputnik moment” claimed Marc Andreesen following their announcement of the open source reasoning model that rivals those made by Silicon Valley tech giants at a fraction of the compute.

Last week, the Chinese startup launched an open source AI assistant that by Monday had become the number one downloaded app on Apple.

Concerns over this breakthrough instigated a $1 Trillion loss in tech stock shares and have reignited the debate over closed versus open systems.

Whether or not this is a Sputnik moment, there are many prognostications surrounding this release. While it is too early to know the long-term impact of this latest shock to the AI world, it is important to both take this breakthrough seriously while also not jumping to the wrong conclusions.

The discourse is certainly going to evolve, but this is not the time for quick conclusions about strategic priorities that will shape the future of AI and geopolitics.

Stargate is not necessary: Many are wondering whether DeepSeek will ‘deep-six’ Stargate, President Trump’s $500 B AI project aimed at developing AI infrastructure in the US. For years, technological bifurcation – the splintering of the physical infrastructure that serves as the backbone of the digital world – has created a divergent system between Chinese backed technologies and those produced by the US and like-minded countries. Governments are moving toward greater data sovereignty and tech sovereignty under the auspices of national security. Stargate is a natural progression of this movement. Given recent attacks on physical infrastructure, including on ViaSat as well as underseas cables, maintaining tech sovereignty over the infrastructure that powers AI is part of the broader global splintering of the internet and technological infrastructure into technospheres. If anything, DeepSeek’s announcement will only deepen this growing divide.
Export controls don’t work: US export controls on the most sophisticated chips inadvertently sparked innovation by requiring Chinese companies to do more with less, or so the argument goes. Export controls will never be perfect, but that does not mean they may not be effective. The array of tech-related export controls have made it much harder for China to develop their own semiconductor industry, and perhaps is better compared to tech containment. Whether its enforcement challenges, or the notion that you can’t ban math – to pull from arguments over encryption bans in recent years – make AI technologies easier to circumvent than other emerging technologies.
Security as an afterthought: With record-breaking downloads, yet again we are witnessing the flight to new tech with security as an afterthought. Nevertheless, DeepSeek had to halt new registrants due to ‘large-scale malicious attacks’. The security risks not only pertain to DeepSeek, but rather include the broad range of attacks, from data poisoning to model corruption. But wait, there’s more. China’s data policies enable the government to access data from companies located in China. Furthermore, given the tight connection between the government and companies, it is naïve to assume complete separation of the Chinese government from DeepSeek going forward, including manipulation and backdoors.
Authoritarian regimes have the innovation edge in AI: While this has not been a prime take-away this week, there has been a growing debate about which regime type has the AI edge, largely based on the greater access to all kinds of data by authoritarian regimes. However, it ignores the censorship and propaganda that can poison the AI models. Existing Chinese GenAI models have already demonstrated censorship and disinformation, and initial research shows DeepSeek’s AI suffers the same problem. Garbage in, garbage out may give democracies the edge, including in reasoning models.

Over a decade ago, Edward Snowden’s revelations helped deepen the divide between Silicon Valley and DC. DeepSeek may finally be the impetus that brings the two group together. Is this the spark that rejuvenates the close allegiance between the tech sector and the US government, similar to Hewlett Packer, Texas Instruments, and IBM from the early days of the Cold War?

Will the AI-focused export controls passed earlier this month target DeepSeek before it becomes a TikTok regulatory problem? If so, how will China retaliate?

The only certainty is that we are still in the early days of this generation-defining technology. AI is more than the technology, but must be viewed through the regulatory, national security, and social systems lens in which it is deeply intertwined.

Go Deeper: What’s in Store for 2025

You can view our take on the 5 trends to look out for in 2025 in our latest report, including our breakdown on the need for Secure AI and the larger Bifurcation of technology along geopolitical fault lines:

Download Now

Interim Final Rule on Artificial Intelligence Diffusion

Posted on January 14, 2025January 15, 2025 by Andrea Little Limbago

Author: Dr. Andrea Little Limbago, SVP, Applied AI, interos.ai

To kick off the New Year, Russian President Vladimir Putin ordered the Russian government and its major bank to coordinate AI development with China. This announcement followed a similar one a few weeks earlier wherein Russia highlighted collaboration among the BRICs (Brazil, Russia, India, and China) and South Africa for an AI alliance.

These announcements, in turn, coincide with a steady drumbeat of AI-driven techno-alliances among the US and its allies, including those between the EU and US, within the QUAD, as well as adjacent policies such as the CHIPs Act and the US AI Executive Order.

Yesterday’s Interim Final Rule on Artificial Intelligence Diffusion is the latest global policy aimed at technological diffusion within allies, which continues to deepen the growing technological bifurcation and upend global supply chains. In the race to implement AI, organizations must stay atop the global technospheres of influence, which will continue to reshape corporate technology stacks or else introduce new security and regulatory risks.

Summary of Bifurcation

The latest wave of AI-focused, technology alliances is a continuation of a pattern that has been going on for years. Technospheres of influence have emerged, wherein part of the world is building upon largely Chinese-created technology infrastructures, and other parts on those built by US and allies.

The US-China trade war initially instigated the nascent splintering almost a decade ago and was followed by US and European export controls and sanctions targeting thousands of companies in China and Russia. China, in turn, has an Unreliable Entity List, which saw the most recent additions on January 2 with the announcement of the addition of ten US defense companies.

These policies have accelerated, both with the increase of geopolitical tensions, but also due to the growing awareness of sanctions circumvention and the use of US-created technology by Russia against Ukraine.

Both the EU and US have specifically targeted distinct rounds of sanctions with anti-sanctions circumvention goals. The result, so far, has been a widening of geographic divisions of technology stacks dependent on geographic location and geopolitical alliances.

Potential Impact

This latest Interim Rule targets foundational AI technologies, including automatic data processing machines, electronic integrated circuits, semiconductors, and calculating machines. The Interim Rule specifically encourages the exchange and research collaboration in these product areas with 18 allies, while restricting their access to ‘non-trusted actors’, a consistent thread among the series of other US AI-related policies over recent years.

Interos.ai identified over 27,000 companies in the US who export these four very specific product categories. These companies, in turn, have global footprints across non-trusted countries and allies alike, as detailed in the table below. Over 20% of companies buying directly from these US companies are in Mexico, followed by India, Great Britain, Colombia, and Canada.

China is among the top 12 direct customers producing one of these products: automatic data processing machines, electronic integrated circuits, semiconductors, and calculating machines. Under the Interim Rule it is assumed that these products could be used in AI technologies.

As you see in the table above, there are thousands of companies who purchase the four product areas listed above. Over 650 of these are in countries of concern, such as China, Russia, and Iran, which exceeds one thousand when looking into the third tier. Meanwhile, almost 4,000 companies are among the 18 allied countries listed in the Interim Rule, and over 3,700 tier 3 companies.

This highlights both the risks and opportunities for companies in complying with the Interim Rule, wherein sizable mats already exist for expansion among like-minded democracies. At the same time, this also illustrates the increasing challenge of doing business in at-risk or adversarial countries.

While these numbers focus on very small, niche product categories, they often are components of much bigger and broader product technology ecosystems.

To that end, when looking at the US tech industry writ large, interos.ai data reveals almost 575,000 companies globally that are directly supplied by a company in the US tech industry. The biggest direct importers from the US technology industry are concentrated in the United Kingdom, India, Australia, Canada, and Mexico.

US AI Policy in Transition

As we noted last Fall, AI governance is critical for shaping the global rules of the road when it comes to AI development, deployment, safety, and security. The EU released the first comprehensive AI policy last year, while the Executive Order and Blueprint for the AI Bill of Rights are the most comprehensive frameworks from the US, but lack the regulatory teeth.

In addition, as often occurs with leadership transitions, there is uncertainty surrounding how the next administration will approach AI. The AI Executive Order is expected to encounter additional scrutiny, with potentially getting repealed based on comments made by the incoming Trump administration. However, based on an AI executive order late in 2020, there are likely areas of continuity as well, indicating that AI policy will remain a moving target.

Geopolitical Tensions will be Central to the Shifting AI Regulatory Landscape in 2025

Given the fast pace and broad impact of AI, the only certainty around the global AI regulatory landscape is that there will continue to be shifts and changes, with geopolitical considerations central to these changes. While the new Interim Rule is the latest example of AI-driven governance updates, it will not be the last.

The geopolitical landscape will continue to drive technological bifurcation, creating distinct technospheres of influence among the US and allies in contrast to China and like-minded regimes.

In addition, we can expect to see changes in AI policies focused on enhancing the underlying security and safety fundamentals of AI. AI security concerns are likely to come front and center in 2025.

On the security front, these will focus on minimizing adversarial AI, including prompt injection attacks, data poisoning, and model manipulation. There also are safety concerns, and we can expect the use cases of specific AI to drive regulatory practices, with higher safety use cases attracting greater regulatory oversight compared to low risk use cases.

The first two weeks of 2025 have already proven eventful for the global AI regulatory landscape. With AI proving to be a generational technology, not only is technological innovation critical, but so too are the governing frameworks surrounding it.

interos.ai views secure AI as a growing and critical consideration for supply chain, full of both opportunities and challenges. We work closely with our customers, supporting their AI governance frameworks and serving as strategic partners to guide AI governance decisions.

To learn more about it and other major trends for 2025, download the interos.ai 2025 Predictions Report.

Download Now

The Race Is on to Shape AI Governance and Security

Posted on November 8, 2024December 16, 2024 by Andrea Little Limbago

Author: Andrea Little Limbago, PhD, SVP, Applied AI

The Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence (EO 14110) was released one year ago. The recent Memorandum on AI builds upon the executive order and focuses on the national security implications of AI, including innovation and leadership within a secure AI framework. At Interos, we take AI very seriously, from building a secure AI framework to launching new AI products, AI is the center point in everything we do.

Artificial Intelligence: The Stakes Could Not be Higher

As the Memorandum details, the timing is critical, as the world undergoes a massive paradigm shift with technological transitions accompanied by global geopolitical shifts.

In the big race to integrate AI, organizations must understand that along with the enormous innovation potential, security and geopolitical considerations cannot be an afterthought.

This Memorandum aims to catalyze change toward a Secure AI framework that supports innovation and leadership, while protecting against adversarial misuse and harm. The stakes could not be higher.

What’s at Stake: Innovation, Economic Growth and Democracy or Authoritarianism and Suppression

Amidst the ongoing AI hype cycle and trillions in investments, it may be easy to forget that AI – like most technologies – is dual-use in nature.

That is, AI can foster innovations and significant breakthroughs, while also enabling more nefarious intentions. As the Memorandum articulates, AI is powering authoritarianism, including malicious cyber behavior, censorship and human rights violations. China is emerging as an ‘AI-tocracy’, using the technology to suppress dissent and entrench regime power. Russia’s notorious bot farms are powered by AI to spread disinformation globally. Iran is similarly deploying AI for influence operations, as well as domestic surveillance and human rights violations.

But AI is also a tool to counter digital authoritarianism. Across the globe, AI is used to pursue democratic values, including empowering political communication, circumventing authoritarian regimes, and heightening defenses against malicious cyber activity. These are just a few examples to underscore the national security imperative detailed in the Memorandum.

The global leader in AI governance will play a critical role in tilting the balance of AI applications toward innovation, economic growth, and democracy, or toward authoritarianism and suppression.

The AI First-Mover Advantage

Strategic competition is front and center throughout the recent Memorandum.

Technology does not exist in a vacuum; the current geopolitical shifts and spread of digital authoritarianism elevate the necessity for the United States to expand its technological edge in this era-defining technology.

Implicit within the Memorandum is that the international order is at an inflection point; the future will not look like the past.

In these situations, first-mover advantage is critical as countries that have garnered the power of breakthrough general purpose technologies gain hegemonic influence in shaping the global order to their advantage.

While the AI technological edge is critical to this, AI governance leadership too often takes a backseat to it. Leadership in AI governance is critical to gaining the first-mover advantage.

Currently, the European Union (EU)’s AI Act is the first major imitative to introduce AI regulations and guardrails. China has also introduced several rules targeting AI, such as the use of generative AI quickly following the release of ChatGPT, but it has yet to formulate a comprehensive AI regulation.

While the US has non-binding AI governance guidelines, such as EO 14110, a comprehensive federal AI regulation does not yet exist. To fill this void, in the 2024 legislative session, 45 states introduced AI legislation, and 31 adopted resolutions or passed legislation.

Last week’s Memorandum clearly identifies the stakes at play, and continues the drumbeat of AI guidance, including the 2022 Blueprint for an AI Bill of Rights.

The US private sector is moving ahead absent a federal framework, introducing AI governance policies at a faster pace than the public sector. The race is on to shape AI governance, and the Memorandum outlines the national security implications for the US to lead this effort, and a partnership across the public and private sectors is critical to solidifying this edge.

Partnership and Collaboration: Protecting the AI Supply Chain

The Memorandum details a whole-of-society approach toward AI. Specifically, the Memorandum contends, “If the United States Government does not act with responsible speed and in partnership with industry, civil society, and academia to make use of AI capabilities in service of the national security mission — and to ensure the safety, security, and trustworthiness of American AI innovation writ large — it risks losing ground to strategic competitors.”

This partnership is critical. While the Memorandum aims to ‘catalyze change’ in how the US government addresses AI national security policy, a similar revolution is necessary in how industry, civil society, and academia approach AI.

Several critical components of the Memorandum directly impact the private sector, such as building and retaining top AI workforce talent, defending against foreign interference and cyber threats, and integrating secure AI in critical infrastructure.

Interos similarly advocates for a Secure AI framework; supply chains and national security are intricately intertwined. This has been made very clear with the Hezbollah device attacks, which marked an inflection point in modern warfare.

According to Interos data, the average enterprise in the S&P 500 has 1,700 direct suppliers and 1.5 million relationships through its first 3 tiers of suppliers. 99% of those companies have ties with at-risk or restricted entities. While the Hezbollah device attacks were not via a restricted company, those technology companies on restricted lists represent a more probable pathway to hardware infiltration and warrant heightened alert – illustrating the widespread vulnerabilities that could be within an organization’s supply chain.

Interos works closely with our customers, supporting their AI governance frameworks and serving as strategic partners to guide AI governance decisions. Secure AI is front in center of our development decisions as well, understanding that different forms of AI introduce different risks, and taking those into account to optimize the implementation of AI coupled with security.

From jailbreaking to data poisoning to algorithmic manipulation, just as supply chains must be secured, so too must the AI supply chain be protected across inputs to algorithms to outputs.

Innovation and security must go hand in hand to truly leverage the vast potential of AI, while protecting ourselves and our supply chains from the growing range of national security risks.

Toward a Secure AI Framework

AI is an era-defining technology. Authoritarian regimes and adversaries are adopting AI at a rapid pace, introducing significant national security threats, including military advantage, global influence, and technological advantage. US leadership is necessary to tip the AI balance toward scientific breakthroughs that support humanity, protect democracy, and empower innovation.

In the race toward AI adoption, security must be at the forefront, not an afterthought.

The world is changing fast; previous paradigms are ill-prepared for ensuring the safety, security, and trustworthiness of our organizations, and our supply chains. AI is both the means toward achieving greater national security, but also poses a great threat if we fail to prepare for its malicious use.

Even without malicious intent, AI systems require greater protection. The latest Memorandum is another critical step toward advancing US leadership in AI, but more is needed.

The public and private sectors alike must internalize the national security imperative at stake or risk ceding this once-in-a-generation technology to the competition.

AI-Powered Supply Chain Risk Management

At Interos, we take AI very seriously. As a global leader in AI-powered supply chain risk intelligence, we are leveraging the power of AI to revolutionize supply chain resilience at a time when global disruptions are at an all-time high.

We recently launched our latest AI innovation, “Ask Interos” that enables organizations to identify supplier threats in real time. It is our first step towards contextual AI. The launch comes at a crucial time when organizations are inundated with data yet struggle to separate complex supply chain noise from actionable insights.

See how we are using AI to secure supply chains in real time.

Get in touch