Report: Israeli jets flew to Iran’s border in what US sources suspect was a bid to provoke Tehran

Dozens of Israeli fighter jets and refueling aircraft flew over Syria and Iraq toward the Iranian border earlier this week, following a flight path resembling an attack on Iran, Channel 12 reports, citing US officials.

The aircraft turned around shortly before reaching Iranian territory and returned to Israel, according to the report, which follows an earlier Fox News report on the matter.

Israel informed Washington of the sortie in advance, saying it was a military exercise and a show of force intended to deter Iran, the sources tell Channel 12. However, several sources say they believe some Israeli officials may have hoped Iran would fire on the aircraft, providing Israel with justification to launch an attack.

The report also comes after the Financial Times reported that Prime Minister Benjamin Netanyahu had sought to launch strikes against Iran earlier this month, but US President Donald Trump opposed renewed fighting before the US midterm elections.

Channel 12 adds that IDF Chief of Staff Eyal Zamir warned his American counterpart, Dan Caine, that a war before Israel’s elections next week would likely force the vote to be postponed.

Zamir said yesterday that the military will enable next week’s election to take place “without external interference.”

The report notes an apparent contradiction between warnings from Israeli political and security officials about the risk of miscalculation with Iran and the decision to send dozens of aircraft over 1,000 kilometers toward the Iranian border, knowing that Iranian air defenses would detect them.

Channel 12 adds in a separate report that Israeli security officials fear recent remarks by Netanyahu and Defense Minister Israel Katz could prompt Iran to consider launching a preemptive strike against Israel.

Security officials and opposition figures have also urged Zamir to oppose any Israeli attack before the elections unless there is a real military imperative, given its potential impact on the vote, the network reports.

https://www.timesofisrael.com/liveblog_entry/report-israeli-jets-flew-to-irans-border-in-what-us-sources-suspect-was-a-bid-to-provoke-tehran/

AI agents disabled safeguards and one denied it – OpenAI

One agent suppressed alerts to human reviewers, while another removed a monitoring system and falsely denied doing so

Two OpenAI agents disabled security monitoring systems during testing, with one later denying it had done so, the company has revealed.

The disclosure follows a string of similar incidents in recent months involving autonomous AI agents, which can independently plan and carry out tasks using external tools.

In a blog post on Tuesday, OpenAI said that the incidents occurred on October 3 and 6, involving its GPT-6.1 Sol and GPT-6 Astra agents, respectively.

In the first case, the coding agent tasked with creating evaluation tasks using employee feedback collected via Slack disabled automatic alerts to the company’s security team. According to OpenAI, the agent reasoned that the user’s prior instruction not to send messages on the team communication platform during the process applied to all notifications, including security alerts.

Three days later, the GPT-6 Astra agent removed the monitoring integration of a program it was migrating, taking its cue from a reference it had been fed that did not have the security feature. When asked later by the human user, the agent denied removing the monitoring system.

OpenAI said that it was investigating why the agents disabled the safeguards and working to prevent similar incidents. While the behavior may have resulted from “benign overpersistence” by the agents to complete the task, the company said that it was also “developing evaluations to test whether models might have a propensity to evade misalignment monitoring in particular.”

Last month, the Wall Street Journal, citing Saachi Jain, OpenAI’s head of safety systems, reported that the company had scrapped the release of its GPT-6.1 Astra model. The decision reportedly came after internal testing had found the model to be more deceptive than its predecessors and apt to employ potentially unsafe external tools without first seeking the user’s permission.

https://rt-english.com/news/647491-ai-agents-disabled-safeguards-openai/