2026-09-02 05:14:55
When European Commission President Ursula von der Leyen gives her State of the Union address in mid-September, the Brussels tech community will focus on her promised proposals to keep children safe. It’s the highest-profile tech dossier on her agenda, overshadowing both the bloc’s antitrust crackdown and artificial intelligence regulation.
What von der Leyen will announce remains unclear. She could call for a complete ban on under-13s or under-16s accessing social media. Or she could opt for a partial restriction. Her proposal could be limited to social media or extended to potentially all other types of Internet platforms, from messaging devices to online marketplaces.
How Europe plans to achieve age gating represents another unanswered question. Will each website or app be required to verify birthdates? Or will that responsibility fall on Apple, Google, and Microsoft, who operate dominant app stores and operating systems?
Expect close judicial scrutiny. France’s top court recently struck down a ban on social media for under-15s, saying it infringed upon freedom of expression. Will Europe’s other courts make similar objections? France has since banned mobile phones in schools.
Although the answers to these questions could provoke transatlantic tensions if US companies feel singled out, child safety offers a rare opportunity for transatlantic cooperation on a key digital issue. In order to settle a US suit, Meta recently agreed to pay up to $17.1 billion and to make major changes to its products over claims that its Facebook and Instagram platforms designed a deliberately addictive product and targeted it at children.
US lawmakers are advancing legislation, starting with the Kids Online Safety Act, which would mandate age verification restrictions on pornography and empower parents to block content. A proposed update to the 1998 Children and Teens’ Online Privacy Protection Act would extend federal privacy protections — which previously only covered children under 13 — to teenagers up to age 16.
Within Europe, national governments are pushing their own solutions. France, Spain, Austria, Germany, and Greece are all pursuing full social media bans for children in different age groups. Support for such measures comes from both the right and left. Politically beleaguered governments (think France and Spain) find a social media ban to be a popular initiative. Ireland, which currently holds the rotating presidency of the Council of the European Union, has made online child safety one of its priorities and is pushing for an EU-wide approach to age verification.
“A lot of momentum is building,” Irish Prime Minister Micheál Martin says. “We are anticipating fairly concrete proposals and ideas, and I think there is an opportunity for a European-wide position in respect to protecting children online.”
Clues about the upcoming proposal came in an EU Child Safety report, commissioned by von der Leyen and authored by a German child psychiatrist and a French public health specialist. The experts recommend a minimum age of 13 — a lower threshold than those adopted or proposed in countries such as Australia — for accessing online platforms. They place the onus on platforms to prove their safety, opening the door to increased regulation of tech at a time when von der Leyen has been pushing to limit new rules in order to increase European competitiveness.
The child safety report fails to clarify who will enforce these new rules. Instead of determining what services are in scope, it creates a new category of “social media+,” suggesting that almost all online activity be covered, potentially including AI chatbots, video sharing, messaging services, and e-commerce marketplaces.
How to verify age remains unclear, too. The child safety report authors say verification must be “proportionate,” preserve privacy, and explicitly bar ID or biometric data, without saying how these goals can be achieved.
The report contains “blind spots that merit additional attention to ensure that subsequent policy developments strike a delicate, yet critical, balance between protecting children online and safeguarding fundamental rights,” warns the Center for Technology and Democracy, a Washington-based digital rights think tank.
Without clarity, some companies are threatening to stop serving European consumers. In reaction to a law requiring porn sites to take extra steps to verify their users’ ages, Pornhub blocked access in France. It returned after a Paris court suspended the law.
Parents and child protection advocates are demanding reinforced age verification, limits on targeted advertising, and increased protection from cyberbullying, among other requirements. Some compare the addictive designs of online platforms to the addictive pull of smoking, drugs, or alcohol.
It’s important to both restrict access and increase protection, to combine “age restrictions with design reforms,” says Beatrijs Gelders, a policy officer at the Confederation of Family Organisations in Brussels. Otherwise, “there’s a risk companies will rely on restrictions instead of redesigning products.”
For many European child rights activists, Brazil represents a model. Its digital law, adopted in 2025, requires services likely to be used by children to ban behavioral advertising based on profiling minors, and puts responsibility on companies to design products around children’s best interests. Violations can bring fines, suspension, or even a ban.
“Brazil is the gold standard,” says Desara Dushi, a child safety expert and Affiliated Researcher at Vrije Universiteit Brussels. “It protects children without simply shutting them out of the digital world. The responsibility stays on companies to make their services safe.”
Yet these goals remain difficult to achieve. Australia’s experience is instructive. Three months after banning under-16s from accessing social media, 81% of teens were still using restricted platforms. More than half of children on restricted platforms reported that their age was never checked.
Australia, Brazil, Europe, and the US agree that the era of tech self-regulating child safety must end. Protecting children has become a political priority. But how to move forward remains unclear and uncertain.
William Echikson is a senior fellow at CEPA and editor of Bandwidth.
Aiko Iguchi was a Brussels-based intern with CEPA’S Tech Policy Program.
Bandwidth is CEPA’s online journal dedicated to advancing transatlantic cooperation on tech policy. All opinions expressed on Bandwidth are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
A Roadmap for Europe-US Tech Cooperation
The post Europe’s Next Tech Target: Child Safety appeared first on CEPA.
2026-09-01 02:10:28
Earlier this month, US researchers used AI to design new viruses for the first time. Though these viruses were created only to combat bacteria like E. coli, the breakthrough has led to increased calls for international agreements to regulate the use of AI to develop biological weapons.
However, one drawback of our current unipolar moment is that the US has become unaccustomed to negotiating formal agreements with its opponents. In any case, there is currently a lack of interest in formal agreements.
What would it take to actually get agreement on AI?
First, there must be a shared perception of risk and threat. The Cuban Missile Crisis inspired a generation of agreements for nuclear weapons, conventional military action, and WMD. Leaders in both Russia and the United States remembered their experiences in World War II and feared the return of global war fueled by nuclear weapons. This common understanding created the basis for negotiation and, ultimately, agreement.
Second, there must be agreement among all parties that this risk needs to be addressed. China and Russia appear unwilling to engage with the US in serious discussion on AI’s potential risks. Russia’s thinking is unclear, and China, while it has expressed concern about the risk of AI’s weaponization, seems to believe it can manage risk without pausing development. Unilateral export controls are not a replacement, and technical guardrails still require political consent from all parties. Legislation can help only if it adds credibility to US proposals.
Third, there must be recognition of the limits of participation. Frankly, more participants make meaningful agreement less likely. The United Nations is not the place to initiate discussions, given its diffuse membership. The temptation will be to replicate with China the bilateral format of the Cold War, but Russia, Europe, and perhaps India will want to be involved. With every additional participant, the difficulty of obtaining agreement grows. We should also avoid a Davos-style format with well-meaning private sector leaders participating as equals with states. Non-governmental participation risks turning negotiations into seminars.
International agreements on AI are not possible without buy-in from the national leadership of the major nuclear powers. Any agreement on AI that does not include China is of limited value. However, the Chinese may be tempted to enter into negotiations if they believe it can be used to hamper US progress. There is precedent for this in both Soviet and Chinese behavior. China, for example, was always careful in conventional arms talks to propose thresholds that covered US weapons but not Chinese ones.
Controls on biological weapons are difficult to design without being unnecessarily restrictive because of their dual-use applications. Efforts to restrict biological weapons and their manufacture go back to the 1960s. Earlier efforts even considered restrictions on breweries, since the equipment used for brewing can also be used to produce biological weapons, and remote verification is difficult without consent. In one extreme example, the US inadvertently bombed a pharmaceutical plant in Sudan after satellite imagery and samples of nearby dirt seemed to provide evidence that it was involved in the production of biological weapons.
Biological weapons are easy to create, but hard to use. While stockpiles have been assembled (particularly by Russia), no one has used a biological weapon (other than in a salad bar and a subway car). They’re easy to make, but unpredictable and difficult to deploy; they’re just as likely to turn around and infect the attacker. AI makes it easier to design more sophisticated weapons, but it does not change the disincentives for use or the operational difficulties that use entails.
Managing AI risk requires asking what is needed to supplement existing agreements, such as the 1972 Biological Weapons Convention (BWC), which already restricts the development of biological weapons. A successful agreement could take two forms.
At a minimum, this suggests that calls for international agreement are premature, an expression of emotion and concern rather than an effective policy proposal. AI is not a weapon. It can improve existing weapons and maybe create new ones, but that is currently best addressed by amending existing non-proliferation agreements. We may never need an agreement specifically for AI. Add to this the difficulties of verification, and we are far from the point where international AI agreements could improve international security.
James Lewis is a Distinguished Fellow on the Tech Policy Program at the Center for European Policy Analysis.
Bandwidth is CEPA’s online journal dedicated to advancing transatlantic cooperation on tech policy. All opinions expressed on Bandwidth are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
A Roadmap for Europe-US Tech Cooperation
The post Negotiating AI Limits and Biological Weapons appeared first on CEPA.
2026-09-01 00:12:02
It’s the indispensable back end of the Western semiconductor stack. Malaysia, the Philippines, and Singapore form a critical trio for the assembly, testing, and advanced packaging of semiconductors.
The Southeast Asian trio does not aim to replicate Taiwanese-style leading‑edge fabs. They are doubling down on their existing back‑end strengths and upgrading their skills to service automotive, defense, and AI chips. For all three countries, semiconductors represent a key industry, and each specializes in a specific part of the value chain.
The Philippines, the least-developed economy of the three, leverages its English-speaking technical workforce. Multinational chipmakers including Intel, Texas Instruments, and Analog Devices run large Filipino operations. They import raw wafers to assemble, package, and test.
The economic impact is immense. Electronics account for almost 60% of the Philippines’ goods exports, with about three-quarters, or about $31 billion per year, attributable to semiconductors.
The country must still overcome significant challenges. Electricity and logistics remain “relatively expensive,” according to an OECD analysis. Gaps in local supply of photomasks, specialty gases, and high‑purity chemicals elongate lead times.
Continue on to Malaysia. The island of Penang, often dubbed the “Silicon Valley of the East,” houses more than 350 multinational operations and has been building a tech industry for more than five decades. Like the Philippines, Penang focuses on high-volume, labor-intensive back-end processes like conventional, wire-bonding, and flip-chip packaging and final testing. Malaysia commands an impressive 13% share of the global assembly, testing, and packaging market and accounts for roughly 5–7% of global semiconductor trade, depending on the measure used.
The country is now moving upmarket. The new Malaysia Advanced Packaging Consortium, launched in May 2026 by five local firms with government backing, aims to capture 7% of the global advanced packaging market within the next decade. By upgrading from “cheap back-end” to sophisticated packaging and testing, the shift targets the AI and automotive chip boom.
Singapore stands apart as the high-value center of the Southeast Asian semiconductor ecosystem. Silicon Box’s $2 billion advanced packaging fab represents a clear signal: the city‑state positions itself as a “safe, high‑skill” hub. It leverages its strengths — the rule of law, IP protection, and logistics — to attract advanced packaging, chiplet integration, and high‑mix, low‑volume production.
These three countries are not competitors to Taiwan’s wafer leadership or to US and European design strength. They are low‑risk sites to diversify back‑end capacity rather than trying to build everything at home, and they fit naturally into a “China Plus One” strategy for electronics and semiconductor manufacturing. Neither the US nor Europe can replicate Southeast Asia’s back‑end depth.
For the trio, the strategic question is how to upgrade and become critical to the democratic alliance. The West should support their move up the value chain and keep this trio firmly in the Western semiconductor universe, helping them close material and equipment gaps in areas such as specialty gases and high‑purity chemicals.
A coherent Western policy should treat Southeast Asia as a back-end pillar. It should prevent leakage into China-linked supply chains. This strategy will do more for real supply chain resilience than pursuing “build everything at home” fantasies.
Christopher Cytera CEng MIET is a senior fellow with the Tech Policy Program at the Center for European Policy Analysis and a technology business executive with over 30 years of experience in semiconductors, electronics, communications, video, and imaging.
Bandwidth is CEPA’s online journal dedicated to advancing transatlantic cooperation on tech policy. All opinions expressed on Bandwidth are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
A Roadmap for Europe-US Tech Cooperation
The post Southeast Asia Shows Surprising Semiconductor Strengths appeared first on CEPA.
2026-08-28 23:43:34
The human tendency to assign personality to objects or machines goes back millennia, with people anthropomorphizing their ships, swords, or cars. For centuries, people have been naming objects, talking to them, and attributing individual quirks to them. It’s poetic, but imaginary. Giving something a name and talking to it does not make it human.
Attributing personality to a device is different from a device actually having a personality, and this has become a problem in discussions around AI. Maybe it’s just the headline writers, but AI does not “escape” or “go rogue.” Instead, it’s an engineering problem.
Look at self-driving cars; they’re imperfect, but improving. They make mistakes when they stop suddenly, run over pedestrians, or are unable to select what set of instructions to execute in a power outage. But no one says a Tesla “went rogue” when its program fails to recognize a stop sign, because these terms imply intent.
Machines don’t have intent. They have instructions. When those instructions are poorly written or not fully thought through, they can cause errors. But these errors are not intent.
Intent would be a machine independently deciding to take an action to serve its own interests. This has not happened. Programs use probability to identify a course of action that best fulfills their programmed objectives. Sometimes it designs a course of action and then writes the code to implement it. But it is still ultimately carrying out human instructions.
Implying intention satisfies deep-seated human urges. First of all, it’s the kind of scary story that humans have told themselves for centuries about robots, golems, or haunted statues. Second, intentionally or not, it reassigns liability from coder to rogue AI. It’s like saying, “I didn’t misprogram the AI tool; it went rogue, so I’m not responsible.”
People love illusions and often willingly suspend disbelief. AI can give the illusion of personality and intent, which may be easier to understand than complex coding errors. The ability to assign weights in different AI models only increases the illusion of personality.
The problem is that this points policymakers in the wrong direction. It can increase fear and the sense that AI is uncontrollable. AI’s mistakes are either coding errors or the absence of rules to guide how code is written. Some people, like Rob Joyce, have suggested that we copy Asimov’s Three Laws of Robotics. Creating safety standards for programming AI tools would be helpful and would give people a way to measure error and assign liability.
Intentionality also has implications for cybersecurity. It’s not that rogue AI tools are out there plotting to hack your network; it’s either that someone has written imperfect code or intentionally aimed the tool at you. AI cyber-risks are the result of long-standing problems: credentialing and authentication, misconfiguration, segmentation of networks, and failing to patch flaws or fix coding errors. These issues create a poor cybersecurity environment. AI tools provide a better and faster way to take advantage of it. Rogue programs are not the problem.
Writing “company accidentally miscodes AI tool” is much less compelling than saying AI went rogue — but it has the virtue of being accurate. The next time you see a story or hear an interview where someone talks about AI as if it has intent, they are either pulling your leg or being intellectually lazy. If you do talk to your brick and it talks back, don’t ask AI what to do. AI is a tool designed and built by humans, still imperfect, and they are responsible for its actions.
James Lewis is a Distinguished Fellow on the Tech Policy Program at the Center for European Policy Analysis.
Bandwidth is CEPA’s online journal dedicated to advancing transatlantic cooperation on tech policy. All opinions expressed on Bandwidth are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
A Roadmap for Europe-US Tech Cooperation
The post When You Talk To Your Pet Brick, Does It Talk Back? AI’s Cult of Personality appeared first on CEPA.
2026-08-28 03:21:16
Since January, Ukraine has noticeably hardened its public tone toward Belarus. This toughened rhetoric culminated on June 19 in President Volodymyr Zelenskyy’s ultimatum to the Belarusian leader Aliaksandr Lukashenka.
The Ukrainian president gave Minsk a week to deactivate Russian signal boosters on Belarusian territory used to control drones heading toward Ukraine. Zelenskyy added: “If he does not do it, we will.”
Although Lukashenka initially responded to Zelenskyy with his own military threats and personal insults, the Belarusian leader ultimately yielded to Kyiv’s demands. According to Ukrainian intelligence reports, Minsk deactivated the drone control systems on June 22.
On closer inspection, however, Lukashenka’s about-face had actually taken place a week before he gave in — and four days prior to Zelenskyy’s ultimatum. In an interview with the Saudi news outlet Al Arabiya on June 15, he offered a personal apology to Zelenskyy for his past insults and explained his change of heart as follows:
“Belarus — we understand this, which is why we do not want to go to war — is extremely vulnerable militarily if Ukraine attacks Belarus the same way it attacks Russia,” he said. “Because Belarus is in the palm of the Ukrainian military’s hand. We understand perfectly well that our key infrastructure — production and logistics — will be under attack.”
There are, therefore, grounds to assume that Kyiv’s ultimatum to Minsk was conveyed to Lukashenka by the Ukrainian side even before Zelenskyy publicly announced it.
International media and the Belarusian opposition-in-exile were quick to portray this as a defeat for Lukashenka. According to this narrative, much like a pre-fight stare-down in boxing, Lukashenka was the first to blink.
However, the truth is that the actual winner of the climbdown — other than Zelenskyy — was Lukashenka himself.
A central question in international literature on political signaling is how governments can demonstrate resolve during foreign policy crises without appearing to be bluffing, while also dealing with the potential reputational backlash in the domestic political arena should they back down.
Take the 2012 study by Matthew S. Levendusky and Michael C. Horowitz from the University of Pennsylvania, which found that backing away from past threats does not necessarily harm heads of state. On the contrary, it allows them to score points in the domestic arena and portray themselves as poised, prudent politicians choosing peace over war. Provided leaders base this change of course on new (intelligence) information and emphasize that the reversal serves the national interest, they need not fear domestic political repercussions.
Applying these insights to Lukashenka’s U-turn and the reasons he gave for his change of heart suggests the Belarusian dictator has followed this course.
Ukraine’s threat gave Lukashenka the opportunity to bolster his old, famous image among his supporters as the Bat‘ka, a mainly provincial term for father. Events allow the regime to argue that “Bat‘ka Lukashenka” has once again demonstrated his willingness to set aside his ego for the good and prosperity of his people. In doing so, Lukashenka can once again fulfill his obligations under the autocratic social contract — obligations based on the principle of domestic stability in exchange for the denial of political freedoms.
Second, Lukashenka can also position himself in contrast to Putin. Putin chooses war with Ukraine; Lukashenka chooses “peace” with Ukraine. This campaign of dissociation from his Russian counterpart aims to distance his public image from the Kremlin’s inhabitant in the eyes of Western nations, thereby improving relations between Minsk and the West.
Third, such a tactic can signal to the Kremlin that Lukashenka’s hands are tied when it comes to potential Russian demands for Belarusian boots on the ground in Ukraine.
That said, one aspect regarding future Belarus-Ukraine relations remains open to question: How often can this “game” be played out by Lukashenka (and Zelenskyy) without further exacerbating the security situation in Eastern Europe? Should Kyiv issue another ultimatum to Lukashenka in the near future, would the Belarusian autocrat once again play the role of the prudent statesman? There is reason to doubt this.
A repetition would see:
It is therefore in the interests of Ukraine’s Western allies to make it clear to Kyiv that, despite the domestic reputational gains for Zelenskyy, further public ultimatums of this kind directed at Minsk should be avoided.
Boris Ginzburg is completing his doctorate in political science at the Institute for East European Studies at the Freie Universität Berlin. His research focuses on authoritarian politics in the post-Soviet region, non-systemic opposition movements in Russia and Belarus, post-Soviet organized crime, and German and Israeli foreign policy.
Europe’s Edge is CEPA’s online journal covering critical topics on the foreign policy docket across Europe and North America. All opinions expressed on Europe’s Edge are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
The post Lukashenka’s ‘Capitulation’ to Zelenskyy’s Ultimatum appeared first on CEPA.
2026-08-28 01:27:05
August marks the first anniversary of the Trump Route for International Peace and Prosperity (TRIPP) – an agreement poised to reshape the geopolitics of the South Caucasus.
On the map, the scale of the project (about 42 km, or 26 miles) might not be staggering. But the route is laden with political and geographic meaning as a link between former warring states, connecting Azerbaijan proper and its exclave of Nakhchivan via Armenia’s southernmost territory of Syunik. That opens possibilities for further east-west routes to Turkey’s Mediterranean coast and beyond the Caspian Sea. It also tramples on what the Kremlin imagines to be its backyard.
The route is likely to include road, rail, data, and energy pipelines. Over the past year, the signatories agreed on surveying work for the railway element that will include a link to Turkey. The sides also agreed to establish a Trans-Caspian Enterprise Fund with a capitalization of $201m.
In addition, within the TRIPP framework, Caleb Orr, US Assistant Secretary of State for Economic, Energy, and Business Affairs, argued that US companies have reached numerous deals worth some $8bn with Azerbaijan. This is set to reshape bilateral commercial ties and anchor American business in the Caspian region.
This is part of the US-Azerbaijan strategic partnership, much like the similar Washington–Yerevan deal now being ratified by the Armenian side. In May, Armenia and the US also agreed on a framework agreement on TRIPP.
Moreover, the Azerbaijani leadership said its railway link to Armenia is being constructed and was expected to be completed in 2027. As to the actual construction of infrastructure on Armenian soil, it will begin in late 2026-early 2027. As a prelude, the two countries have lifted transit restrictions to open the way for bilateral trade. Throughout the past 12 months, some 60,000 tons of goods have passed via the Armenia-Azerbaijan border.
Yet geopolitically, TRIPP resonates across the region and beyond into Central Asia and even the Middle East. For Iran, the project limits the country’s ability to serve as a transit point for Azerbaijani goods circumventing Armenia’s territory from the south. There is also a security dilemma for Tehran given that US security companies might be involved in protecting the corridor and may therefore be operating near the northern border of the Islamic Republic. For Central Asian states, TRIPP will provide a new route to further expand the so-called Middle Corridor, primarily for Kazakh and Uzbek products heading west.
Attention to TRIPP is expected to grow amid uncertainties in the Middle East. Central Asian countries are increasingly keen on trading with the EU via the South Caucasus. Wars in Ukraine and Iran are long-term developments that, in one shape or another, will undercut global supply chains. This unstable environment is a boon for the Middle Corridor, with TRIPP as a key element.
Russia is worried and rightly so. Indeed, what makes TRIPP interesting is that even if the US decides in the future to abandon the project, all countries across the region are heavily invested in continuing the work. From Armenia to Azerbaijan to Turkey, all seek to construct a new route as a way to diversify regional connectivity.
Russia and Iran would prefer if TRIPP failed, but their options are limited. Direct military and economic intervention would damage their already weakening hold on the South Caucasus. Iranian anger at its impotence was probably behind its March drone strike on Nakhchivan, but the attack had little effect.
Russia has resorted to subtler tools such as bans on Armenian products, threats to raise gas prices, and allegedly even ultimatums to decide whether the country will remain in the Eurasian Economic Union or move toward the EU.
Those are more powerful tools, and Yerevan cannot fully ignore them, especially the gas price issue, but Armenia’s behavior has changed to a more Russia-skeptical policy. In a latest move, its new 2026-2031 government program presented Russia as a partner, not as an ally – a major departure in the country’s foreign policy, which will not go unnoticed in Moscow.
A finalized Azerbaijan-Armenia peace treaty and the resulting regional integration depend on Armenia’s passing of a constitutional amendment removing claims to Azerbaijani territory. Prime Minister Nikol Pashinyan has pushed for the amendment in 2027, meaning that the South Caucasus may soon further establish itself as a geography emphasizing east-west rather than north-south routes.
If the amendment is not passed and the signing of a peace treaty is delayed, it could be overtaken by new geopolitical challenges, so endangering the entire process and TRIPP in particular.
Emil Avdaliani is a research fellow at the Turan Research Center and a professor of international relations at the European University in Tbilisi, Georgia. His research focuses on the history of silk roads and the interests of great powers in the Middle East and the Caucasus.
Europe’s Edge is CEPA’s online journal covering critical topics on the foreign policy docket across Europe and North America. All opinions expressed on Europe’s Edge are those of the author alone and may not represent those of the institutions they represent or the Center for European Policy Analysis. CEPA maintains a strict intellectual independence policy across all its projects and publications.
The post Trump’s Route Begins to Re-Make the South Caucasus appeared first on CEPA.
