Security

• Na conferência CETaS em 1º de maio de 2026, o diretor Alexander Babuta destacou os avanços do Claude Mythos Preview da Anthropic em matemática, cibersegurança, engenharia de software e detecção de vulnerabilidades. • O modelo de fronteira promete ferramentas automatizadas aprimoradas para profissionais de segurança. • Especialistas expressam otimismo sobre o potencial defensivo da IA contra ameaças de hacking potencializadas.

• A Câmara dos Representantes dos EUA aprovou um projeto de lei de dotações para o Departamento de Segurança Interna na quinta-feira, abrindo caminho para encerrar uma paralisação de dois meses da agência que havia interrompido partes das operações federais de imigração, fronteira e cibersegurança. • O presidente Donald Trump assinou a medida pouco depois, restaurando o financiamento total para o Departamento de Segurança Interna e evitando possíveis interrupções na segurança dos aeroportos, nas operações de deportação e na fiscalização de fronteiras até o fim do ano fiscal. • O conflito da paralisação colocou aliados de Trump contra republicanos centristas e democratas em relação aos níveis de fiscalização de fronteiras, mas o projeto final preservou iniciativas fundamentais da era Trump na fronteira, ao mesmo tempo em que suavizou algumas exigências de gastos de linha dura.

• Oracle lançou um novo comunicado de segurança para clientes em 30 de abril de 2026, antecipando-se a ameaças a modelos de AI. • O comunicado destaca um prenúncio de cibersegurança, instando a atualização das posturas de segurança para implantações de AI. • Ele sinaliza medidas proativas contra novas vulnerabilidades visando sistemas de AI generativa.

Secretária do Interior promete ‘fazer tudo ao meu alcance’ para manter os judeus britânicos seguros enquanto a polícia afirma que o suspeito é um cidadão britânico nascido na Somália Shabana Mahmood prometeu “fazer tudo ao meu alcance” para manter os judeus britânicos seguros, conforme a polícia informou que o suspeito do esfaqueamento de dois homens em Golders Green, no norte de Londres, na quarta-feira, era um cidadão britânico de 45 anos nascido na Somália. Ministros disseram que outros £25 milhões seriam investidos para aumentar a segurança das comunidades judaicas após o suposto ataque terrorista no norte de Londres. Continue lendo...

Article 42.7 had languished in obscurity for decades – until Donald Trump began casting doubt on US commitment to Nato• Don’t get This Is Europe delivered to your inbox? Sign up hereMost people have heard of Nato’s article 5. The “one for all, all for one” clause states an armed attack on one member country should be considered an attack on all, requiring member states to come to the victim’s aid – including with “the use of armed force”.Not so many, till this week, had heard of the EU’s own mutual defence clause, article 42.7 (pdf), which says that if a member state comes under armed attack, the others “shall have towards it an obligation of aid and assistance by all the means in their power”. That’s perhaps because there hadn’t, until recently, been much need for Europeans to consult article 42.7. More than 40 US military bases and 85,000 troops across the EU (and UK) were testament to Washington’s defence commitment to the old continent. Continue reading...

An 89-year-old man armed with a shotgun wounded an employee at a social security office in Athens before opening fire at a courthouse, injuring several more people. Police are actively searching for the suspect, with the motive behind the attacks currently unknown.

An 89-year-old man was apprehended in Patras after allegedly injuring at least four people in two separate shootings in Athens. Authorities believe pension grievances and psychological issues fueled the attacks, which targeted a social security office and a court building. The suspect, found with a weapon, is under investigation.

Motorists are forced to queue up to six hours for fuel or pay more on the black market, while farmers lack fuel to operate machinery and harvest crops.

The regional bloc agreed to safeguard regional energy security through strengthened cooperation, timely policy responses and engagement with dialogue partners.

The Politburo, a top decision-making body of the ruling Communist Party, was cited as saying by state news agency Xinhua that the economy got off to a better-than-expected start this year.

• GetReal Security appointed cybersecurity veteran Varun Kohli as Chief Marketing Officer to address surging enterprise needs. • The hire responds to rising deepfake and AI-powered identity attacks in the cybersecurity landscape. • Kohli's expertise will drive expansion of real-time protection solutions for businesses facing AI threats.

Fiona Hill tells MPs UK is ‘vulnerable’ because it does not educate people on how to deal with information warfareBritain is becoming a soft target for Russian and other state propaganda because the UK is not prepared to educate people on how to deal with information warfare, according to a former White House adviser and security expert.Fiona Hill told a parliamentary committee that she feared the UK had become “extraordinarily vulnerable” to online manipulation feeding into the electoral system because there was a lack of discussion about civil defence. Continue reading...

• China's National Development and Reform Commission prohibited Meta's acquisition of Singapore-based AI startup Manus on April 27, 2026, requiring all parties to withdraw. • Manus, with Chinese roots, was targeted in a security review of foreign investment despite Meta's compliance claims from its California headquarters. • The decision heightens US-China tech tensions, impacting Meta's AI expansion and cross-border M&A in semiconductors and intelligence tools.

Secret Service director says security succeeded in stopping shooter before he could do further harm but others disagreeThe shooting in the White House correspondents’ gala has prompted questions over security with some asking how a shooter was able to get close to where Donald Trump and many other senior administration officials were gathered and many others praising the actions of law enforcement that swiftly stopped the attack.As details about the shooting at the Washington Hilton continued to surface, the alleged shooter Cole Tomas Allen, 31, mocked an “insane” lack of security at the Washington dinner in a manifesto reportedly send to his family 10 minutes before his assault started. Continue reading...

Widely dispersed wind farms and solar panels are harder to target than fossil fuel power stations, Michael Shanks saysRenewable energy will boost the UK’s national security and make the country more resilient against potential aggression or sabotage, the government’s energy minister has said.Michael Shanks said widely dispersed wind farms and solar panels were much harder to target than large-scale fossil fuel power stations. They are also not vulnerable to supply shocks, such as the current oil crisis caused by the US-Israel war on Iran and the soaring gas prices that followed Russia’s invasion of Ukraine in 2022. Continue reading...

UK cabinet minister says there are extensive discussions over security after gunman attempted to storm dinner attended by TrumpUS politics live – latest updatesKing Charles’s security is being reviewed before his state visit to the US this week after a gunman attempted to storm a dinner with Donald Trump in Washington DC, a UK cabinet minister has said.Guests at the White House correspondents’ dinner on Saturday night hid under tables when gunshots were heard as the president and other members of his administration were evacuated by the Secret Service. Continue reading...

Extra social media checks brought in amid growing threat to politicians from extremistsThe security company that provides bodyguards for MPs has tightened its vetting processes after it sent a bodyguard with far-right links to protect a politician who was under threat from extremists.Mitie, which has a £31m contract for the work, is updating its CPO (Close Protection Operative) vetting processes to include regular social media checks. There will also be random checks on the social media activity of those already taken on. Continue reading...

• The UN Security Council failed to reach consensus on a resolution extending cross-border humanitarian aid to Syria, with Russia vetoing the measure on Friday amid escalating tensions between Moscow and Western nations over the Syrian conflict's humanitarian toll. • Russia's veto marks the third such blocking in two years, preventing critical medical supplies and food assistance from reaching 5.5 million internally displaced Syrians, according to UN humanitarian coordinator statements. • Western diplomats warned the blockade could exacerbate an already severe humanitarian crisis, with aid agencies reporting shortages of vaccines, antibiotics, and nutrition programs across northern Syria.

• Google Cloud Next 2026 concluded with more than 250 announcements on products, customers, and ecosystem developments focused on AI and enterprise tools. • Key security features include Agent Anomaly Detection using LLMs to flag unusual AI agent behavior in real time and Chrome Enterprise's AI-aware extension threat detections now in preview. • Chrome Enterprise will soon offer shadow AI reporting to identify unsanctioned web-based AI and SaaS use by employees, enhancing visibility into shadow AI risks.

• The First Trust NASDAQ Cybersecurity ETF (CIBR) experienced a pullback in recent trading sessions amid market volatility. • Investors assess buy-the-dip opportunities as the ETF tracks key US cybersecurity firms. • RAD Intel, a holding, scaled from $10 million to over $220 million valuation since acquiring its AI engine in 2021, achieving 5,400% growth.

• Copperhelm raised $7 million in funding to launch an agentic cloud security platform leveraging AI agents for threat detection. • The platform automates real-time remediation at scale, addressing escalating cloud vulnerabilities in US enterprises. • This funding enables Copperhelm to commercialize innovative AI-driven cybersecurity amid rising breach incidents.

• OpenAI and Microsoft expanded their cybersecurity partnership to jointly combat emerging cyber threats through deployment of advanced artificial intelligence tools and enhanced security protocols. • Microsoft will leverage its cybersecurity expertise to strengthen protections for OpenAI's systems and customer environments, while OpenAI's powerful AI models will be made accessible to Microsoft's security operations. • The collaboration aims to address rising AI-driven cyber risks and establish industry standards for AI security defense.

• OpenAI and Microsoft announced an expanded cybersecurity partnership aimed at combating emerging cyber threats in the AI era. • The two companies plan to deploy advanced artificial intelligence tools to enhance security capabilities across their platforms. • This collaboration reflects the growing recognition that AI-powered security solutions are essential to address increasingly sophisticated cyber threats.

• Cyware was again named to the Cyber66 list of hottest privately held security companies, announced via PRNewswire on April 23, 2026. • The recognition highlights Cyware's leadership in AI-powered threat intelligence operationalization and secure threat sharing. • Company emphasizes agentic AI for accelerating targeted threat responses and collaboration.

Report from Elizabeth Warren calls Trump administration cuts to Social Security Administration ‘catastrophic’Sign up for the Breaking News US email to get newsletter alerts in your inboxCuts to the Social Security Administration have caused “customer service chaos” for millions of older Americans and those with disabilities who rely on the agency’s services, according to a new report from a group of Democratic senators.An investigation found that phone wait times were more than 10 times higher than what the agency claimed on its website, if the calls were even answered at all. Continue reading...

• IBM announced IBM Autonomous Security on April 15, 2026, deploying coordinated AI agents to detect and contain threats at machine speed without human intervention. • The service addresses a critical gap: attackers now achieve full network lateral movement in as little as 27 seconds, compared to an average of 29 minutes in 2024—a 65% speed improvement driving urgent demand for AI-powered defenses. • AI-enabled attacks surged 89% year-over-year in 2025, with a 44% increase in assaults exploiting public-facing applications, prompting major security vendors to adopt autonomous AI defense strategies.

• Anthropic is investigating unauthorized access to its Mythos AI tool, a critical cybersecurity system, after reports revealed that an unauthorized group breached the platform via a vendor vulnerability. • The incident raises fresh concerns about security gaps within advanced AI systems and amplifies questions about the trustworthiness of AI tools handling sensitive security functions. • The breach exemplifies broader risks in the interconnected tech ecosystem, where vendor compromises can cascade into exposures of high-value AI infrastructure.

• Google Cloud and Wiz demonstrated integrated security platforms at Google Cloud Next '26 designed to defend against threats operating at machine speed while protecting AI and multicloud environments. • The partnership addresses the convergence of AI deployment, cloud infrastructure expansion, and sophisticated cyber threats, offering coordinated threat detection and response capabilities across distributed architectures. • The showcase reflects industry-wide recognition that traditional security models are obsolete against AI-accelerated attacks and that cloud providers must embed security-by-design into AI and multicloud infrastructure.