Cybercrime

Anthropic uncovers large-scale AI model theft operations

Three AI laboratories have been found conducting large-scale illicit campaigns to extract capabilities from Anthropic’s Claude AI, the company revealed.

DeepSeek, Moonshot, and MiniMax used around 24,000 fraudulent accounts to generate more than 16 million interactions, violating terms of service and regional access restrictions. The technique, called distillation, trains a weaker model on outputs from a stronger one, speeding AI development.

Distilled models obtained in this manner often lack critical safeguards, creating serious national security concerns. Without protections, these capabilities could be integrated into military, intelligence, surveillance, or cyber operations, potentially by authoritarian governments.

The attacks also undermine export controls designed to preserve the competitive edge of US AI technology and could give a misleading impression of foreign labs’ independent AI progress.

Each lab followed coordinated playbooks using proxy networks and large-scale automated prompts to target specific capabilities such as agentic reasoning, coding, and tool use.

Anthropic attributed the campaigns using request metadata, infrastructure indicators, and corroborating observations from industry partners. The investigation detailed how distillation attacks operate from data generation to model launch.

In response, Anthropic has strengthened detection systems, implemented stricter access controls, shared intelligence with other labs and authorities, and introduced countermeasures to reduce the effectiveness of illicit distillation.

The company emphasises that addressing these attacks will require coordinated action across the AI industry, cloud providers, and policymakers to protect frontier AI capabilities.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

AWS warns of AI powered cybercrime

Amazon Web Services has revealed that a Russian-speaking threat actor used commercial AI tools to compromise more than 600 FortiGate firewalls across 55 countries. AWS described the campaign as an AI-powered assembly line for cybercrime.

According to AWS, the attacker relied on exposed management ports and weak single-factor credentials rather than exploiting software vulnerabilities. The campaign targeted FortiGate devices globally and focused on harvesting credentials and configuration data.

AWS said the potentially Russian group appeared unsophisticated but achieved scale through AI-assisted mass scanning and automation. When encountering stronger defences, the attackers reportedly shifted to easier targets rather than persist.

The company advised organisations using FortiGate appliances to secure management interfaces, change default credentials and enforce complex passwords. Amazon said it was not compromised during the campaign.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

Wikipedia removes Archive.today links

Wikipedia editors have voted to remove all links to Archive.today, citing allegations that the web archive was involved in a distributed denial of service attack.

Editors said Archive.today, which also operates under domains such as archive.is and archive.ph, should not be linked because it allegedly used visitors’ browsers to target blogger Jani Patokallio. The site has also been accused of altering archived pages, raising concerns about reliability.

Archive.today had previously been blacklisted in 2013 before being reinstated in 2016. Wikipedia’s latest guidance calls for replacing Archive.today links with original sources or alternative archives such as the Wayback Machine.

The apparent owner of Archive.today denied wrongdoing in posts linked from the site and suggested the controversy had been exaggerated. Wikipedia editors nevertheless concluded that readers should not be directed to a service facing such allegations.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

OpenClaw exploits spark a major security alert

A wave of coordinated attacks has targeted OpenClaw, the autonomous AI framework that gained rapid popularity after its release in January.

Multiple hacking groups have taken advantage of severe vulnerabilities to steal API keys, extract persistent memory data, and push information-stealing malware instead of leaving the platform’s expanding user base unharmed.

Security analysts have linked more than 30,000 compromised instances to campaigns that intercept messages and deploy malicious payloads through channels such as Telegram.

Much of the damage stems from flaws such as the Remote Code Execution vulnerability CVE-2026-25253, supply chain poisoning, and exposed administrative interfaces. Early attacks centred on the ‘ClawHavoc’ campaign, which disguised malware as legitimate installation tools.

Users who downloaded these scripts inadvertently installed stealers capable of full compromise, enabling attackers to move laterally across enterprise systems instead of being confined to a single device.

Further incidents emerged on the OpenClaw marketplace, where backdoored ‘skills’ were published from accounts that appeared reliable. These updates executed remote commands that allowed attackers to siphon OAuth tokens, passwords, and API keys in real time.

A Shodan scan later identified more than 312,000 OpenClaw instances running on a default port with little or no protection, while honeypots recorded hostile activity within minutes of appearing online.

Security researchers argue that the surge in attacks marks a decisive moment for autonomous AI frameworks. As organisations experiment with agents capable of independent decision-making, the absence of security-by-design safeguards is creating opportunities for organised threat groups.

Flare’s advisory urges companies to secure credentials and isolate AI workloads instead of relying on default configurations that expose high-privilege systems to the internet.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!  

Data breach at PayPal prompts password resets and transaction refunds

PayPal has notified some customers of a data breach linked to its Working Capital loan application, after unauthorised access between 1 July and 12 December 2025 exposed personal information. Letters dated 10 February confirm that around 100 customers were potentially affected.

The incident was linked to an error in the Working Capital application, described as a ‘code change’. PayPal said it ‘terminated the unauthorised access to PayPal’s systems’ after discovery.

In a statement sent following publication, a PayPal spokesperson said ‘When there is a potential exposure of customer information, PayPal is required to notify affected customers. In this case, PayPal’s systems were not compromised. As such, we contacted the approximately 100 customers who were potentially impacted to provide awareness on this matter.’

Data potentially accessed includes names, email addresses, phone numbers, business addresses, Social Security numbers, and dates of birth. PayPal confirmed a small number of unauthorised transactions and said refunds were issued. Affected users had passwords reset and were offered credit monitoring.

Previous incidents include a 2023 credential stuffing attack that affected nearly 35,000 accounts and phishing campaigns that abused legitimate infrastructure. The company said it continues to use manual investigations and automated tools to mitigate fraud.

Customers are advised to use unique passwords, avoid unsolicited links, verify urgent messages directly via their accounts, and enable passkeys where available. Even limited breaches can heighten risks of targeted phishing and identity theft, especially for small businesses.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Cloudflare outage causes global internet disruption after an internal error

A major outage on 20 February disrupted global internet traffic after an internal configuration failure at Cloudflare caused the unintended withdrawal of customer BGP routes.

The incident lasted just over six hours and left numerous services unreachable, despite early fears of a cyberattack. An internal update led to the systematic deletion of more than a thousand Bring Your Own IP prefixes, which pushed many connections into BGP path hunting instead of stable routing.

Engineers traced the disruption to an error in the company’s Addressing API, introduced during an automated cleanup task under the Code Orange resilience programme.

A flawed query interpreted an empty value as an instruction to delete all returned prefixes, removing essential bindings for hundreds of customers. Some users restored connectivity through the dashboard, while others required manual reconstruction carried out across the edge network.

An outage that affected a series of core offerings, including content delivery, security layers, dedicated egress and network protection services. Restoration took several hours because the withdrawn prefixes varied in severity, demanding different recovery methods instead of a uniform reinstatement process.

The error triggered widespread timeouts on dependent websites and applications, along with 403 responses on the 1.1.1.1 DNS resolver.

Cloudflare plans to introduce stricter API validation, circuit breakers for abnormal deletion patterns, and improved configuration separation. It has also issued a public apology for a failure that undermined its assurances of network resilience.

An event that reaffirmed the risks posed by internal automation faults when they interact with critical internet infrastructure.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!  

Phishing messages target IndiaAI and Impact Summit 2026 participants

IndiaAI has issued an urgent advisory warning of a phishing campaign targeting attendees of the India AI Impact Summit 2026. Fraudulent SMS and WhatsApp messages claim refunds are pending and request sensitive financial details.

Organisers said the messages are not official and have not been authorised. Recipients are being urged to click links and provide full card numbers, WhatsApp numbers, and other contact information to ‘process’ refunds.

IndiaAI advised participants not to click suspicious links or share personal or banking information with unverified sources. Attendees in India are encouraged to delete such messages immediately and block the sender’s number.

Anyone who may have submitted details through a suspicious link should contact their bank without delay to secure their accounts. Organisers stressed that event-related communication will only be shared through official channels.

The advisory was issued under the AI Impact Summit 2026 banner, themed ‘Welfare for All | Happiness of All’, as authorities seek to prevent financial fraud linked to the high-profile gathering.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Strict ban on crypto references introduced by OpenClaw

OpenClaw has introduced a firm community rule prohibiting any reference to Bitcoin or other cryptocurrencies on its Discord server, according to its creator, Peter Steinberger.

Enforcement drew attention after a user was removed for mentioning Bitcoin block height as a timing method in a benchmark, with the developer later offering to restore access.

The policy follows a rebrand scare when scammers hijacked old accounts to promote a fake Solana token. Market value spiked then plunged after Steinberger denied involvement, warning that no official token would be issued.

Rapid growth of the open-source project, which has attracted a large developer base within weeks of launch, contrasts with wider industry momentum linking AI agents and digital assets.

Leaders such as Jeremy Allaire of Circle argue stablecoins could become default payment rails for autonomous software, while Coinbase is already rolling out infrastructure enabling agents to transact on-chain.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot! 

Generative AI presents the biggest data-risk challenge in history

Cybersecurity specialists warn that generative AI systems, such as large language models, are creating a data risk frontier far larger than that posed by previous digital innovations.

Because these models are trained on extensive datasets drawn from web pages, internal documents, email corpora and proprietary sources, they can unintentionally memorise or regenerate sensitive information, increasing the risk of exposure.

The article highlights several core concerns. Data leakage and memorisation, where AI models can repeat or infer private data if training processes are not tightly controlled.

Amplification of poor hygiene, when generative tools can magnify the reach of bad actors by automating phishing, social engineering, and malware generation at scale.

Compounding breach impact, if an AI model is trained on stolen or leaked data, it could internalise and regurgitate that information without detection, entrenching harm. Cloud and access governance gaps that allow organisations to adopt AI without robust access controls and encryption may widen their attack surface.

The author calls for revised data governance frameworks, including strict training data provenance, auditability, encryption, minimisation and purpose limitation, to mitigate what is described as ‘the biggest data risk in history.’

Recommendations also include accountability measures for models, continuous monitoring, and legislative action to align AI development with privacy and security principles.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Fake Google Forms phishing campaign targets job seekers

A phishing campaign is targeting job seekers with fake Google Forms pages designed to harvest account credentials. Attackers are using a spoofed domain, forms.google.ss-o[.]com, to mimic the legitimate Google Forms service and trick victims into signing in.

The fraudulent pages advertise a Customer Support Executive role and prompt applicants to enter personal details before clicking a ‘Sign in’ button. Victims are then redirected to id-v4[.]com/generation.php, a domain previously linked to credential harvesting campaigns.

Researchers identified the operation as part of a broader wave of job-themed phishing attacks. The attackers used a script called generation_form.php to create personalised tracking links and implemented redirects to evade security analysis by sending suspicious visitors to local Google search pages.

Security experts warn that the campaign relies on domain impersonation techniques, including the use of ‘ss-o’ to resemble ‘single sign-on’. The fake site reproduces Google branding elements and standard disclaimers to increase credibility.

Users are advised to avoid clicking unsolicited job links, verify opportunities through official channels, and enable multi-factor authentication. Password managers and real-time anti-malware tools can also reduce exposure to credential theft.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Diplo chatbot can make mistakes. Consider checking important information.