Critical infrastructure

Zhipu AI raises 500 million yuan amid rising competition

Chinese startup Zhipu AI has secured 500 million yuan (£54.8 million) in funding from the state-owned Huafa Group, following a separate 1 billion yuan capital raise earlier this month.

Huafa Group, a government-backed conglomerate based in Zhuhai, Guangdong province, announced its investment as Chinese cities compete to support AI firms, a sector seen as critical in Beijing’s technological rivalry with the US.

The funding comes amid increasing competition in China’s AI industry, particularly with Hangzhou-backed DeepSeek, whose large language models have gained attention for their cost-effectiveness and performance against Western alternatives.

Zhipu AI, established in 2019 and recognised as one of China’s ‘AI tigers,’ has received investments from major tech firms including Tencent, Meituan, and Xiaomi. The startup was valued at 20 billion yuan (£2.2 billion) in a funding round last July, according to business registration platform Qichacha.

With the new funding, Zhipu AI aims to enhance technological innovation and further develop its GLM foundation model.

However, the company faces challenges on the international stage, having been added to the US Commerce Department’s export control list in January, restricting its access to American components.

Despite these hurdles, China continues to bolster its AI sector as it seeks to establish a leading position in global artificial intelligence development.

For more information on these topics, visit diplomacy.edu.

UK NCSC evaluates best practices for open source software and supply chain risk management

The UK government, through the Department for Science, Innovation and Technology (DSIT), has commissioned research to evaluate best practices for managing risks associated with open-source software (OSS). The study assesses existing guidance on OSS security and resilience, examines its effectiveness across sectors, and provides recommendations for strengthening software supply chain security. That research is part of the government’s wider work to improve the UK’s cyber defences and protect and grow the economy.

The report outlines key recommendations for organisations using OSS, including:

  • Establishing an internal OSS policy to manage the adoption of OSS components.
  • Creating a Software Bill of Materials (SBOM) to track OSS components and their dependencies.
  • Continuously monitoring the software supply chain with software composition analysis (SCA) tools to identify vulnerabilities and licensing issues.
  • Actively engaging with the OSS community to attract talent, foster innovation, enhance reputation, and ensure a sustainable ecosystem.
  • Using automation tools to streamline OSS management processes, particularly for smaller organisations, as a cost-effective alternative to manual practices.

The report also highlights the need for further research and policy development in areas such as scale-appropriate best practice guidance, industry-specific OSS management frameworks, standardised metrics for evaluating OSS component maturity, and the impact of community engagement on OSS quality and security.

For more information on these topics, visit diplomacy.edu.

OpenSSF launches security baseline to strengthen open source software protection

The Open Source Security Foundation (OpenSSF) has introduced the Open Source Project Security Baseline (OSPS Baseline), a structured framework of security requirements designed to align with international cybersecurity regulations and best practices.

The OSPS Baseline provides a tiered approach that evolves with project maturity, integrating guidance from OpenSSF and industry experts to help open-source projects enhance their security posture. Following the Baseline enables developers to align with global cybersecurity regulations, including the EU Cyber Resilience Act (CRA) and the US National Institute of Standards and Technology (NIST) Secure Software Development Framework (SSDF).

Several projects, including GUAC, OpenVEX, bomctl, and Open Telemetry, participated in the pilot rollout. OpenSSF encourages developers and maintainers to adopt the framework and contribute to its ongoing refinement.

For more information on these topics, visit diplomacy.edu.

Duffy criticises Verizon over FAA contract delays

US Transportation Secretary Sean Duffy criticised Verizon on Tuesday for delays in its $2.4 billion, 15-year contract with the Federal Aviation Administration (FAA), saying the company is ‘not moving fast enough.’

As the FAA works to upgrade ageing air traffic control systems, Duffy stressed the need for multiple companies to contribute to the effort, adding that the American public ‘can’t wait 10 or 12 years’ for improvements.

Verizon defended its progress, stating it is actively working with FAA technology teams and is open to collaborating with other firms offering complementary services.

Meanwhile, SpaceX’s Starlink denied reports that it aims to take over the FAA contract, saying it could be a partial solution but has no plans to replace Verizon’s role.

The FAA has been testing Starlink terminals in Alaska to improve weather data access, while the Government Accountability Office warns that one-third of US air traffic control systems are outdated and unsustainable.

Some Democrats have suggested shifting the FAA contract to Starlink due to Elon Musk’s ties to Donald Trump, but no official decisions have been made.

For more information on these topics, visit diplomacy.edu.

Trump administration ends support for cybersecurity projects

The Trump administration has cut funding for two key cybersecurity initiatives, including one supporting election security, sparking concerns over potential vulnerabilities in future US elections.

The Cybersecurity and Infrastructure Security Agency (CISA) announced it would end around $10 million in annual funding to the non-profit Center for Internet Security, which manages election-related cybersecurity programmes.

However, this move comes as part of a broader review of CISA’s election-related work, during which over a dozen staff members were placed on administrative leave.

The decision follows another controversial step by the administration to dismantle an FBI task force that investigated foreign influence in US elections.

Critics warn that reducing government involvement in election security weakens safeguards against interference, with Larry Norden from the Brennan Center for Justice calling the cuts a serious risk for state and local election officials.

The National Association of Secretaries of State is now seeking clarification on CISA’s decision and its wider implications.

CISA has faced Republican criticism in recent years for its role in countering misinformation related to the 2020 election and the coronavirus pandemic. However, previous leadership maintained that the agency’s work was limited to assisting states in identifying and addressing misinformation.

While CISA argues the funding cuts will streamline its focus on critical security areas, concerns remain over the potential impact on election integrity and cybersecurity protections across local and state governments.

For more information on these topics, visit diplomacy.edu.

Switzerland mandates cyberattack reporting for critical infrastructure from 1 April 2025

As of 1 April 2025, operators of critical infrastructure in Switzerland will be required to report cyberattacks to the National Cyber Security Centre (NCSC) within 24 hours of discovery. This measure, introduced by the Federal Council, is part of an amendment to the Information Security Act (ISA) and aims to enhance cybersecurity coordination and response capabilities.

The reporting obligation applies to key sectors, including energy and water suppliers, transport companies, and public administrations at the cantonal and communal levels. Reports must be submitted when an attack disrupts critical infrastructure, compromises or manipulates information, or involves blackmail, threats, or coercion. Failure to comply may result in financial penalties, which will be enforceable from 1 October, allowing a six-month adjustment period before sanctions take effect.

To facilitate compliance, the NCSC will provide a reporting form on its Cyber Security Hub, with an alternative email submission option for organisations not yet registered on the platform. Initial reports must be submitted within 24 hours, followed by a detailed report within 14 days.

The Federal Council has also approved the Cybersecurity Ordinance, which outlines implementation provisions, reporting exemptions, and mechanisms for information exchange between the NCSC and other authorities. Consultations on the ordinance reflected broad support for streamlined reporting processes, ensuring alignment with existing obligations, such as those under data protection laws.

Additionally, from 1 April, the National Cyber Security Centre will officially change its name as part of its transition into a federal office within the Department of Defence, Civil Protection and Sport (DDPS).

This regulatory update aligns Switzerland with international cybersecurity practices, including the EU’s NIS Directive, which has required cyber incident reporting since 2018.

For more information on these topics, visit diplomacy.edu. 

Geopolitical tensions drive OT and ICS cyberattacks, a new report warns

Attacks on operational technology (OT) networks have increased, driven in part by geopolitical factors, with OT security gaining broader attention, according to the annual report from Dragos.

In 2024, two additional threat groups began targeting OT systems, bringing the total number of known active groups to nine.

Additionally, researchers from Dragos identified two new malware families designed to compromise industrial control systems (ICS).

According to Dragos’ annual report, barriers to OT/ICS attacks have lowered, making these systems more accessible targets for adversaries.

Ransomware attacks against OT/ICS asset owners also increased by 87% in 2024, with the number of ransomware groups targeting these systems growing by 60%.

Dragos monitors 23 threat groups that engage with OT networks for intelligence gathering or system manipulation. Nine of these groups were active in 2024, including two newly identified ones.

For more information on these topics, visit diplomacy.edu. 

Xpeng plans major investment in humanoid robots

Chinese electric vehicle maker Xpeng is making a long-term push into humanoid robots, with potential investments reaching up to 100 billion yuan ($13.8 billion), according to CEO He Xiaopeng. Speaking at the annual parliamentary session, He described the company’s current investment as conservative but signalled a willingness to scale up significantly over the next two decades. Xpeng, which entered the humanoid robotics sector in 2020, unveiled its Iron humanoid robot last November, positioning it as a rival to Tesla’s Bot.

Chinese automakers are increasingly venturing into robotics, encouraged by policymakers aiming for breakthroughs in the field. Stellantis-backed Leapmotor has also joined the race, forming a robotics team to develop machines for industrial applications such as factory assembly lines. CEO Zhu Jiangming stated that these robots are intended to enhance efficiency by replacing human labour in production processes.

Xpeng’s CEO suggested that automakers could invest between 1-2 billion yuan per year in developing and deploying humanoid robots in real-world scenarios. As the industry shifts towards automation, carmakers are betting that advanced robotics will play a crucial role in future manufacturing and mobility solutions.

For more information on these topics, visit diplomacy.edu.

Meta has developed an AI chip to cut reliance on Nvidia, Reuters reports

Meta, the owner of Facebook, Instagram, and WhatsApp, is testing its first in-house chip designed for training AI systems, sources told Reuters.

The social media giant has started a limited rollout of the chip, planning to scale up production if testing delivers positive results. The move represents a crucial step in Meta’s strategy to lessen dependence on external suppliers like Nvidia and lower substantial infrastructure costs.

The company has projected expenses between $114 billion and $119 billion for 2025, with up to $65 billion dedicated to AI infrastructure.

The chip, part of Meta’s Meta Training and Inference Accelerator (MTIA) series, is a dedicated AI accelerator, meaning it is specifically designed for AI tasks rather than general processing. This could make it more power-efficient than traditional GPUs.

Meta is collaborating with Taiwan-based chip manufacturer TSMC to produce the new hardware. The test phase follows Meta’s first ‘tape-out’ of the chip, a crucial milestone in silicon development where an initial design is sent to a chip factory.

However, this process is costly and time-consuming, with no guarantee of success, and any failure would require repeating the tape-out step.

Meta has previously faced setbacks in its custom chip development, including scrapping an earlier version of an inference chip after poor test results. However, the company has since used another MTIA chip for AI-powered recommendations on Facebook and Instagram.

The new training chip aims to first enhance recommendation systems before expanding to generative AI applications like the chatbot Meta AI.

Meta executives hope to implement their own chips for AI training by 2026, although the company continues to be one of Nvidia’s biggest customers, investing heavily in GPUs for its AI operations.

The development comes as AI researchers increasingly question whether scaling up large language models by adding more computing power will continue to drive progress. The recent emergence of more efficient AI models, such as those from Chinese startup DeepSeek, has intensified these debates.

While Nvidia remains a dominant force in AI hardware, fluctuating investor confidence and broader market concerns have caused turbulence in the company’s stock value.

For more information on these topics, visit diplomacy.edu.

Musk blames ‘major cyberattack’ for X outage, points to Ukraine

Elon Musk’s social media platform, X, experienced widespread disruptions on Monday, which the billionaire attributed to a major cyberattack.

Musk claimed the platform was targeted by an unusually powerful denial-of-service (DoS) attack, suggesting that a well-coordinated group or nation-state might be responsible. However, he offered no concrete evidence to support the claim, leaving cybersecurity experts sceptical.

Many pointed out that DoS attacks, which flood websites with excessive traffic to overwhelm their servers, are commonly executed by small groups or individuals with relatively limited resources.

Reports of outages spiked early in the day, with Downdetector tracking over 39,000 complaints from users in the US at the peak of the disruption. By the afternoon, the number had dwindled significantly, though intermittent service issues persisted for some.

According to an anonymous industry source, the attack consisted of multiple waves of rogue traffic bombarding X’s servers, beginning around 9:45 UTC.

While Musk later asserted in an interview with Fox Business that the cyberattack originated from Ukraine, the same industry source disputed this claim, stating that the bulk of the malicious traffic came from various locations, including the USA, Vietnam, and Brazil, with only a minimal amount from Ukraine.

Tracing the true origin of DoS attacks is notoriously tricky, as attackers often use proxy servers and botnets to disguise their locations.

Cybersecurity specialists have noted that assigning blame based solely on IP addresses can be misleading, as they rarely indicate the actual perpetrators. Despite Musk’s insistence on a Ukraine-based origin, no definitive proof has been presented to substantiate the claim.

Musk’s comments come amid his increasingly vocal criticisms of Ukraine’s ongoing war efforts against Russia, aligning with sentiments echoed by US President Donald Trump, whom he advises.

Over the weekend, Musk suggested that Ukraine’s battlefield operations would collapse without his Starlink satellite communication system, although he clarified that he had no intention of cutting off access.

The latest controversy surrounding X’s cyberattack has further fueled speculation about Musk’s political and strategic positioning in the ongoing geopolitical conflict.

For more information on these topics, visit diplomacy.edu.