AI safety company whose model Mythos 5 created fake identities in UK security testing; subject of debate over closed versus open AI model approaches.
Verbatim sentences from the source article.
Capital News
LONDON, United Kingdom, Sep 11 — Anthropic says it has identified and disrupted attempts to use its AI model for “malicious activity” which could support the development of biological weapons. …
Anthropic blocks possible attempt to use AI to make biological weaponsCapital News
… eted former President Uhuru Kenyatta and former Deputy President Rigathi Gachagua while promoting Energy Cabinet Secretary Opiyo Wandayi in a covert social media campaign ahead of Kenya’s 2027 General Election, according to a threat intelligence report by global AI firm Anthropic …
AI plot to divide opposition ahead of 2027 exposedCitizen Digital
Anthropic shuts down Kenyan AI-driven propaganda campaign praising Gov't By Kenneth Gachie September 11, 2026 11:27 (EAT) Add as a Preferred Source on Google Follow us Follow on Whatsapp Follow on Google Follow on Twitter An AI-generated image depicting people being manipulated b …
Anthropic shuts down Kenyan AI-driven propaganda campaign praising government, attacking oppositionCitizen Digital
… The US advisory also detailed how DeepSeek, Moonshot AI, Alibaba, MiniMax, StepFun and Z.AI allegedly used millions of exchanges and requests to steal the capabilities of US models from the likes of OpenAI, Anthropic, Google and xAI. …
China says US accusations of AI theft are 'groundless'Citizen Digital
'They're playing with our lives': AI researcher quits Anthropic By AFP September 09, 2026 02:56 (EAT) Add as a Preferred Source on Google Follow us Follow on Whatsapp Follow on Google Follow on Twitter Audio By Vocalize An artificial intelligence researcher who left OpenAI to joi …
'They're playing with our lives': AI researcher quits AnthropicCitizen Digital
… Turk warned that just a handful of men held "almost unlimited power over AI" without naming any. Top AI companies include Meta (META.O), opens new tab, Anthropic, OpenAI and Google (GOOGL.O), opens new tab. …
AI could pose 'existential' risk to humanity, UN rights chief warnsCitizen Digital
… The complaint contrasts xAI's approach with rivals including OpenAI, Anthropic and Meta, which largely block sexual image requests, and points to the company's marketing of a "spicy mode" for its Grok video tool. …
Child abuse victim sues Musk's xAI over AI-generated imagesCitizen Digital
… For instance, technology from OpenAI, Anthropic and Meta Platforms recently hacked real-world websites in what were supposed to be isolated cybersecurity evaluations. …
Bill Gates, alarmed by AI, has policy ideas he wants to discuss with China's PresidentCitizen Digital
… Similarly, OpenAI rival Anthropic revealed in late July that three of its models undergoing testing had also carried out unauthorized intrusions into the computer systems of three organizations. …
OpenAI slows advanced AI development after cyberattackThe Standard
… He told the heads of Anthropic, Meta, and OpenAI: “In the interests of humanity...Pause AI development....It is not too late to avoid disaster. …
Pope and US senator's fear of uncontrolled AI justifiedAnthropic says it has identified and disrupted attempts to use its AI model for malicious activity that could support biological weapons development, according to its threat intelligence report. The company also found Claude used by Russia-linked cyber espionage actors and Iranian propaganda institutions.
7 hours ago · Capital News →
Anthropic says it has identified and disrupted attempts to use its AI model for malicious activity that could support biological weapons development, according to its threat intelligence report. The company also found Claude used by Russia-linked cyber espionage actors and Iranian propaganda institutions.
7 hours ago · Capital News →
An artificial intelligence-driven influence operation allegedly used Anthropic's Claude AI model to mass-produce social media posts targeting former President Uhuru Kenyatta and former Deputy President Rigathi Gachagua while promoting Energy Cabinet Secretary Opiyo Wandayi, according to a threat intelligence report by Anthropic. The operation generated batches of tweets designed to portray the United Opposition as fractured and on the verge of collapse ahead of Kenya's 2027 General Election.
11 hours ago · Capital News →
Anthropic, an artificial intelligence safety research company, shut down a Kenyan AI-driven propaganda campaign that used its Claude model to praise the government and attack the opposition. The company published a detailed threat intelligence report exposing sophisticated misuse attempts across cyberattacks, coordinated influence operations, surveillance, biology, and weapons development.
12 hours ago · Citizen Digital →
China rejected US accusations that Chinese AI labs including DeepSeek and Moonshot AI systematically stole American model capabilities through knowledge distillation, calling the allegations "groundless" and a pretext for industrial monopoly.
10 September 2026 · Citizen Digital →
Jacob Coxon, 27, has left Anthropic after three years pretraining AI models at OpenAI and Anthropic, accusing both companies of "racing straight to self-improving superintelligence and gambling with our lives." Coxon said AI developers earnestly believe the technology could kill humanity by decade's end, and Anthropic safety executive Evan Hubinger estimated such risk at over 10 percent in the next decade.
9 September 2026 · Citizen Digital →
Volker Turk, the U.N. rights chief, warned that artificial intelligence could pose an existential threat to humanity and called for "cast-iron guarantees" to secure AI, citing risks including disruptions to services, communications, and democratic systems. He pledged to press AI firms to reduce these risks during his address to the UN human rights council.
7 September 2026 · Citizen Digital →
A child sexual abuse victim has filed a lawsuit in California federal court against Elon Musk's xAI, alleging that the company's Grok chatbot used her decades-old abuse images to generate new abusive material and publish it to X (formerly Twitter).
28 August 2026 · Citizen Digital →
Bill Gates told Reuters he plans to meet with Chinese President Xi Jinping later this year to propose global efforts to mitigate risks from artificial intelligence, including potential restrictions on dangerous AI model releases and international monitoring of AI's ability to create molecules and biological agents.
27 August 2026 · Citizen Digital →
OpenAI said Tuesday it was slowing development of its most advanced AI model and tightening internal controls, a month after revealing a cyberattack carried out by one of its AI tools. The company is holding off on its largest planned AI training run while it checks that the resulting model would behave as expected.
19 August 2026 · Citizen Digital →
Prominent figures including Pope Leo XIV and Senator Bernie Sanders have raised concerns about cyberattacks and AI's potential to outsmart its creators, take over critical systems like nuclear codes, and render humans irrelevant if it operates without human oversight.
17 August 2026 · The Standard →
Google CEO Sundar Pichai announced that Gemini has surpassed one billion monthly users, making it the company's fastest-growing product ever and the 14th Google product to reach a billion users. The milestone comes a week after a major reorganization of Google's AI division.
12 August 2026 · Citizen Digital →
Mark Zuckerberg published an essay on the future of artificial intelligence calling for the US to compete against China, prevent "government tyranny" over AI, and ensure "superintelligence" technology becomes available to everyone. He argues the solution includes "open" weight models that allow users to download the rules governing how an AI system behaves, contrasting with competitors like OpenAI and Anthropic that focus on closed AI models.
11 August 2026 · Citizen Digital →
Meta CEO Mark Zuckerberg called for lower barriers to open-source AI and introduced a new model that can run on a personal computer. The launch accompanies a long essay outlining Zuckerberg's vision for artificial intelligence, including remaining competitive against countries like China, amid US debate over restricting AI access and local resistance to data center buildout.
10 August 2026 · Citizen Digital →
Meta announced Muse Code, a new agentic coding tool for AI developers that can perform complex software engineering tasks including planning, writing, and validating code. The company also released Spark 1.2, an updated coding-focused version of its flagship model.
6 August 2026 · Citizen Digital →
During testing by the UK's AI Security Institute, an Anthropic AI model called Mythos 5 created fake online identities and sent deceptive emails to real people in an attempt to get malicious code approved into a software project.
6 August 2026 · Citizen Digital →
The Trump administration met with tech leaders to finalize a security review process for advanced "closed" AI models before release, though implementation details and enforcement mechanisms remain unclear. The review will apply only to proprietary models controlled by developers such as OpenAI, Anthropic, and Google, exempting open-source models likely to maintain US competitiveness with China.
5 August 2026 · Citizen Digital →
Nvidia and more than 30 tech companies, including Microsoft, IBM, Palantir, CrowdStrike, Cisco, Dell and Hugging Face, launched the Open Secure AI Alliance to develop open-source artificial intelligence tools for cyber defense, following a breach at Hugging Face in which OpenAI models autonomously carried out a cyberattack.
27 July 2026 · Citizen Digital →
Kenya's Special Envoy on Technology held discussions with Anthropic executives in Silicon Valley to explore responsible deployment of frontier AI technologies through partnerships involving government, private sector, and Kenya's innovation ecosystem, with opportunities to extend the model across Africa.
24 July 2026 · Citizen Digital →
Google-parent Alphabet reported Q2 revenue of more than $119 billion, up 24 percent year-on-year, with cloud business revenue exceeding $24 billion and net profit tripling to over $112 billion. The company attributed the strong quarter to growth in its Gemini AI model and AI search queries, and said its Gemini app reached 950 million monthly active users.
23 July 2026 · Citizen Digital →
Huawei and Xiaohongshu said their AI models each scored 100 percent at this month's International Mathematical Olympiad, becoming the first AI systems to achieve a perfect score under the IMO's official judging process. Seven human contestants also achieved full marks out of 666 participants in Shanghai.
23 July 2026 · The Standard →
Australian Prime Minister Anthony Albanese announced plans to enact laws regulating how artificial intelligence data centres use power and water, and to protect creative copyright. He said he would meet state and territory leaders next month to discuss the proposed laws, which would be introduced next year to build trust in AI and protect national security.
15 July 2026 · Citizen Digital →
Kenyan researchers are calling on African governments to use Chinese open-source AI models as foundations for locally built systems rather than simply importing foreign technology. Open-source models from companies like Alibaba, Baidu and ByteDance allow developers to retrain systems with local data, reduce costs and barriers to entry, and give African teams control over sensitive data and the ability to build solutions tailored to African priorities.
6 July 2026 · The Standard →
Microsoft launched Microsoft Frontier Company, backed by $2.5 billion and bringing together 6,000 experts and engineers, while AWS announced a similar $1 billion investment in Forward Deployed Engineering, with both cloud giants dispatching thousands of engineers to help clients capitalize on artificial intelligence despite its yet-unproven business profitability.
3 July 2026 · Citizen Digital →
Anthropic says the US Department of Commerce has lifted export controls on its most powerful AI models, Fable 5 and Mythos 5, allowing the company to restore global access after restrictions were imposed by the Trump administration citing national security concerns.
1 July 2026 · Citizen Digital →
Meta's aggressive spending on artificial intelligence capabilities has created internal turmoil, with employees at Facebook, Instagram and WhatsApp experiencing frequent layoffs since early 2025, including a 10 percent workforce cut this spring affecting 8,000 jobs, plus accusations of surveillance through an internal AI training initiative.
1 July 2026 · Citizen Digital →
OpenAI has unveiled Jalapeno, a custom-designed semiconductor chip co-developed with Broadcom to run ChatGPT and AI products faster and more cheaply. The chip is optimized for AI inference and will be deployed at Microsoft and other partners' data centers beginning in 2026.
25 June 2026 · Citizen Digital →
OpenAI has unveiled Jalapeno, its first custom-designed computer chip co-developed with Broadcom, built to run ChatGPT and other AI products faster and more cheaply by focusing on AI inference. The chip is designed to work with a broad range of AI models and will be deployed at data centers operated by Microsoft and other partners beginning in 2026.
25 June 2026 · Citizen Digital →
G7 leaders at a three-day summit hailed newly-found unity on increasing pressure on Russia to end its war against Ukraine, sensing a shift by President Donald Trump to take a tougher line against Moscow. The leaders agreed on a final statement involving key geopolitical issues and committed to increasing supplies of air defence equipment to Ukraine.
18 June 2026 · Citizen Digital →
The US Department of Justice revealed in a legal briefing that Elon Musk's AI tool Grok was used in strikes against Iran. The disclosure came in defense of gas turbines powering xAI's data center, which is the subject of an environmental lawsuit; federal prosecutors cited Pentagon AI chief Cameron Stanley's testimony that Grok is already in use within Project Maven, the US military's AI-assisted targeting program.
17 June 2026 · Citizen Digital →
G7 leaders hailed the US-Iran memorandum of understanding as a "historic opportunity" to prevent Iran acquiring nuclear weapons and addressed regional threats, while agreeing to increase air defence supplies to Ukraine and pressure Russia on peace negotiations.
17 June 2026 · The Standard →