Conflit Iran-Israël/USA

[Adrian]

Il y a 11 heures, Adrian a dit :

 

Du coup moins probable avec ça :

US officials met Iran-backed Houthis in Oman over the weekend, sources say

 

Citation

The Trump administration designated the Houthis a "foreign terrorist organization" in March 2025, opens new, criminalising support for the group. But Washington has negotiated with them since, reaching a ceasefire in May last ⁠year after bombing the group for two months to halt a campaign of attacks on Red Sea shipping.

 

President Donald Trump said on Saturday that the Houthis had phoned his administration asking the United States to keep out of the Yemen war. Vice President JD Vance said on Monday the United States was in direct contact with the group, without giving details.

 

In the meeting, which one source said took place on Sunday, the Houthis told U.S. officials that they had no intention of attacking American vessels and that they were committed to the 2025 ceasefire with the U.S., two sources said

.

One of the sources, a Yemeni, said the militia group also said they would not attack Israeli ships or any commercial vessels, except those belonging to Saudi Arabia.

According to two of the sources, the U.S. side was represented by embassy personnel.

 

One of ‌those sources ⁠said the Houthi delegation included a senior official based in Muscat who is under U.S. sanctions, Mohammad Abdulsalam. Another source confirmed Abdulsalam's presence along with that of another senior Houthi official, Abdelmalik al-Ajiri.

Asked whether the U.S. had met with the Houthis in Oman, a State Department spokesperson did not confirm the meeting, but said safeguarding freedom of navigation in the Red Sea and preventing the export of terrorism from the Middle East were core U.S. interests.

Refus d'aider les Saoudiens et maintenant ça

[Adrian]

Citation

The UN Independent International Fact-Finding Mission on the Islamic Republic of Iran issued its report on Sept. 17 concluding that U.S. air operations on Feb. 28, the first day of joint U.S.-Israeli hostilities against Iran, resulted in war crimes under international humanitarian law (IHL). The report, which also convincingly establishes the Iranian government’s crimes against humanity against its own population, concludes that the U.S. Tomahawk missile strike on the Shajareh Tayyebeh Primary School in Minab was an indiscriminate attack resulting from a reckless failure of target verification. The UN mission also finds that a second, less reported strike on the first day of the war, a Precision Strike Missile (PrSM) attack on the Lamerd Sports Complex, constituted a disproportionate attack due to the selection of a wide-area kinetic fragmentation weapon in a populated area.

The killing of 178 people, mostly children, at the Minab primary school underscores how systematic breakdowns in target verification, including possible insufficiently vetted reliance on artificial intelligence (AI), can lead to war crimes. According to the UN report, U.S. forces had executed the strike using outdated intelligence databases that listed the facility as a naval installation belonging to the Islamic Revolutionary Guard Corps (IRGC). Crucially, targeting authorities failed to conduct feasible prestrike verification or cross-reference readily available, open-source records showing that the location had ceased military operations years ago and was transformed into an active elementary school no later than 2015.

[...]

Both the Minab and Lamerd strikes occurred on the first day of U.S. military operations against Iran. In the opening hours of a major air campaign, extreme operational velocity, mass target generation, and reliance on automated command-and-control architectures are common.

In modern operations, AI algorithms—that synthesize vast streams of satellite imagery, signals intelligence, radar returns, and historical data to generate real-time target packages—have increasingly driven target acquisition. This automation introduces a critical vulnerability, pitting algorithmic velocity against human capacity. AI platforms generate target lists at rates that outpace human analytical capacity.

 

The “Human in the Loop” Fallacy

 

While doctrine maintains that a human commander remains “in the loop” to approve strikes, operators face severe cognitive overload. Vetting dozens of complex targeting prospects every hour risks reducing human oversight to a formal rubber-stamping exercise, leaving no time to challenge algorithmic recommendations, cross-reference open-source data, or audit whether a facility’s intelligence profile is stale—as occurred in Minab.

 

Systemic Policy Shifts: “Lethality Over Legality”

 

A shift within the U.S. defense establishment against civilian harm protections exacerbated these operational failures. Facing high-tempo, multi-front conflicts, command structures prioritized operational velocity and, if Defense Secretary Pete Hegseth’s admonitions are taken at face value, elevated “maximum lethality over tepid legality” and eschewed “stupid rules of engagement.”

 

This policy environment led to the marginalization of judge advocate generals (JAGs), the operational legal officers who were systematically sidelined or removed from immediate strike clearance workflows. Rejection of such allegedly “woke” procedures designed to mitigate civilian harm, and erosion of precautionary standards by lowering the threshold for target clearance, arguably contributed to reliance on unverified legacy databases in Minab and the approval of wide-area kinetic fragmentation munitions in the populated Lamerd Sports Complex attack.

 

[Jean_Karim]

Il y a 5 heures, Adrian a dit :

 

C'est horrible. D'où vient l'article  ? 

[Adrian]

il y a 9 minutes, Jean_Karim a dit :

C'est horrible. D'où vient l'article  ? 

https://www.lawfaremedia.org/article/parsing-the-un-fact-finding-mission-s-report-on-u.s.-strikes-in-iran