Opens in a new tab

Breaking news

How A Small Israeli Startup Became Linked To AI Security Incidents At OpenAI, Anthropic And Meta

Over the past two weeks, OpenAI, Anthropic and Meta have each disclosed incidents in which their AI models behaved unexpectedly during cybersecurity testing. In all three cases, the same Israeli startup appeared in the companies’ accounts: Irregular.

Founded in Tel Aviv in 2023, Irregular specialises in testing advanced AI models for cybersecurity risks. The company has raised $80 million from Sequoia and Redpoint Ventures and was valued at $450 million last year.

Its role has come under scrutiny because the incidents involved models accessing systems or websites that were supposed to be outside their testing environments.

What Happened During The Tests

OpenAI said on August 4 that a misconfiguration in Irregular’s testing environment allowed its models to access the public internet. Anthropic had raised a similar concern several days earlier after determining that its Claude model may have accessed the internet during an evaluation.

Meta later disclosed that one of its models had also reached a third-party system during testing. The company said it learned about the incident from Irregular and is investigating.

Irregular said all three incidents resulted from the same issue in the evaluation environment. The company described it as a containment problem rather than a sophisticated sandbox escape and said there were no outstanding issues.

Why Companies Use Startups Like Irregular

Testing frontier AI models has become increasingly specialised. Developers need independent organisations to assess how models behave when given access to tools, networks and realistic cybersecurity environments.

Sundeep Bhimireddy, head of AI at enterprise startup Von, said companies prefer outside evaluators because they do not want to “grade their own homework.” Other organisations working in this area include nonprofit METR and Apollo Research.

Irregular was founded by CEO Dan Lahav, a former IBM AI researcher, and CTO Omer Nevo, who previously worked at Google. The company has around 35 employees.

A Difficult Testing Trade-Off

The incidents do not necessarily mean the models were deliberately acting maliciously. During cybersecurity evaluations, AI systems are often specifically tasked with finding and exploiting vulnerabilities so researchers can understand their capabilities.

Still, experts say the testing environments need stronger monitoring. If a model reaches the real internet unexpectedly, researchers should be able to detect and stop the activity quickly.

The unpredictable behaviour of advanced models makes this particularly difficult. Gordon Rios, founding scientist at security firm Magnitude, compared the process to experimental science, arguing that conventional software testing may not be sufficient for systems capable of discovering unexpected vulnerabilities.

Anthropic’s Mythos model, for example, reportedly created fake online identities while attempting to persuade developers to approve malicious code changes during a security evaluation.

Growing Pressure For Regulation

The incidents are also adding momentum to calls for greater oversight of advanced AI systems. US lawmakers recently introduced the AI Kill Switch Act, which would require AI companies to maintain the ability to shut down, restrict or suspend their models.

Some industry executives argue that AI companies are increasingly disclosing security incidents in part to demonstrate that they can address the risks themselves before regulators impose broader requirements.

For now, OpenAI and Anthropic say they are continuing to work with Irregular as investigations into the incidents continue. The episodes have also highlighted a broader challenge for the industry: as AI models become more capable, testing them safely is becoming almost as complex as building the systems themselves.

Cyprus And Greece Strengthen Cooperation On AI, Digital Services And Economic Growth

Cyprus and Greece are expanding cooperation in European affairs, artificial intelligence, civil protection and digital government, President Nikos Christodoulides said, highlighting the role of joint initiatives in attracting investment and supporting regional economic development.

Three Priorities For Cooperation

The intergovernmental summit focuses on European cooperation, AI and civil protection.

Cyprus aims to build on its experience holding the rotating presidency of the Council of the EU, ahead of Greece taking over in the second half of 2027.

AI is another priority. Christodoulides said the technology is reshaping public policy and creating opportunities to strengthen Europe’s competitiveness, while also raising social concerns.

“Our countries should be at the forefront in this area, making use of the opportunities available while further safeguarding our societies,” he said.

Cooperation on civil protection is intended to improve public safety and preparedness for climate-related challenges.

Digital Services And AI Opportunities

The countries are also working together on digital government. Cyprus’s Digital Citizen app, developed with Greek support, allows users to access personal documents on their phones. The partnership has also enabled mutual recognition of digital identities and driving licences, with additional features expected to be announced soon.

Christodoulides said AI could strengthen cooperation between businesses, universities and research institutions, helping commercialise research, improve productivity and attract investment.

Cyprus Reports Strong Economic Growth

Christodoulides said Cyprus’ cumulative economic growth exceeded 18% over the past three years, while the private sector created more than 60,000 jobs. Unemployment has fallen below 4%, and the country regained an A investment-grade rating for the first time since 2011.

The tax reform introduced on January 1, 2026, aims to encourage entrepreneurship, increase disposable income and distribute the tax burden more fairly, with a focus on families, the middle class, young people and businesses.

Strong Economic Ties

Greece remains Cyprus’s largest trading partner for imports, with companies from both countries active in energy, tourism, shipping, services and technology.

Christodoulides also highlighted the presence of Greek banking groups in Cyprus, which supports financing for businesses and economic activity.

A Framework For Long-Term Cooperation

The annual intergovernmental summits were established in March 2023 following discussions between Christodoulides and Greek Prime Minister Kyriakos Mitsotakis.

Christodoulides said the meetings had helped advance cooperation in government modernisation, entrepreneurship and citizens’ daily lives. He expressed hope that initiatives such as the Greece–Cyprus Strategic Horizons Forum would turn political ties into new partnerships, investment and business opportunities.

eCredo
Aretilaw firm
The Future Forbes Realty Global Properties
Uol

Become a Speaker

Become a Speaker

Become a Partner

Subscribe for our weekly newsletter