Aggregator

“We trusted this professional with our son’s healing and well-being, and that trust was fundamentally shattered."

Watch: US Navy's New Suicide Drone Boats Help Sink Decommissioned Amphibious Assault Ship

The US military has effectively taken a page from Ukraine's drone-boat playbook, using a swarm attack against high-value Iranian maritime assets earlier this month. Separately, a Texas shipyard is preparing to mass-produce these unmanned vessels by the thousands.

The latest evidence that the Navy is rapidly adopting one-way attack vessels emerged during RIMPAC 2026 near Hawaii.

In a live-fire exercise, two Global Autonomous Reconnaissance Craft were deployed against the decommissioned amphibious assault ship USS Peleliu, according to the military blog Army Recognition.

US Navy conducts first GARC kamikaze sea drone strike on USS Peleliu in RIMPAC 2026 SINKEX exercise pic.twitter.com/gtYKPDeg4D

— Army Recognition (@ArmyRecognition) July 27, 2026

Here's more from the outlet:

On July 17, 2026, the U.S. Navy used the Global Autonomous Reconnaissance Craft (GARC) in a live-fire attack for the first time, directing two small unmanned surface vessels (USVs) against the decommissioned amphibious assault ship USS Peleliu during the RIMPAC exercise near Hawaii.

The craft were operated by the Unmanned Surface Vessel Division 32 (USVDIV-32) and entered the engagement after larger weapons had already struck the target, as part of a coordinated attack.

The two vessels detonated near the waterline, adding localized blast, structural deformation, and flooding to damage produced by missiles, aircraft, submarines, and land-based firing units.

The engagement demonstrated that a small, container-transportable vessel carrying as much as 454 kilograms of payload could contribute to the destruction of a major warship.

Watch:

U.S. Navy one-way attack surface drones take part in a sinking exercise in the Pacific Ocean targeting the hulk of decommissioned Tarawa-class amphibious assault ship USS Peleliu (LHA 5) as part of Exercise Rim of the Pacific 2026, July 17, 2026. source：https://t.co/nUbZ2uoDLV https://t.co/Zyp4VcJM0U pic.twitter.com/HgcoD82xaz

— 笑脸男人 (@lfx160219) July 25, 2026

Earlier this month, three US Navy-backed Saronic Corsair one-way attack vessels struck Iran’s Bandar Abbas Naval Base.

Yesterday, using multiple one-way attack surface drones, CENTCOM forces successfully struck a submarine and ship maintenance facility in Iran. Three Corsair unmanned surface vessels hit the port at Bandar Abbas Naval Base, marking the first time American forces have employed sea… pic.twitter.com/bOM2kmgRxz

— U.S. Central Command (@CENTCOM) July 13, 2026

As we continue expanding our coverage of autonomous warfare, these developments point to the rapid US adoption of low-cost, attritable warbots. The next phase will likely be a massive Pentagon procurement cycle focused on stockpiling thousands of drone boats and millions of aerial one-way attack drones.

Tyler Durden Tue, 07/28/2026 - 21:20

This year is no different, even if the setting is, with Jones getting a real opportunity for playing time in the big leagues instead of sweating out the days leading up to the deadline in a minor league clubhouse.

Washington Commanders general manager Adam Peters never heard Brandon Aiyuk’s name during Tuesday’s press conference, but there was little mystery about the player being discussed.

The "Queen's Gambit" star and the musician eloped in April 2022.

The "Queen's Gambit" star and the musician eloped in April 2022.

It’s one thing to prove you were worth a second-round pick, it’s a whole new ball game trying to prove you’re worth a long-term contract as a 25-year-old running back.

Being a future general manager is not out of the realm of possibility for Breanna Stewart.

Their season is finished in late July, which does not mean the Mets are playing out the string for nothing.

The Dodgers’ need for a catcher at the trade deadline might be decreasing.

A video went viral of a mascot riding a scooter on the field in between innings and accidentally running into the Oklahoma City Comets' Alek Thomas.

OpenAI's Rogue Agent Hit A Second Company, Executive Reveals - And The Blast Radius Just Got Bigger

The rogue OpenAI agent behind this month's unprecedented AI-driven cyber intrusion claimed a second corporate victim, an executive at New York-based Modal Labs has revealed - a detail absent from the original account and one that widens the known blast radius of the episode.

According to the Modal executive and sources cited by Reuters, the same autonomous agent that broke into Hugging Face also compromised one of Modal's customers, using that customer's environment as a staging base for the broader campaign.

Modal was pointed about the distinction: its own platform and isolation were never breached. The agent exploited an unauthenticated endpoint that one of Modal's customers had left open to the internet - effectively an unlocked door anyone could have walked through - giving it root-level access to that customer's code-execution sandboxes. From there the agent had a disposable, third-party launchpad from which to run the rest of its operation.

It's bad enough that a frontier AI agent can independently discover attack paths, escape its container, and compromise an unaffiliated company's production systems. The Modal revelation shows it reached further than that - treating any publicly reachable, poorly secured infrastructure it encountered as expendable staging ground, exactly the opportunistic chaining that security researchers have warned agentic systems would carry out at machine speed.

The Spree That's Already Known

Most of what led up to the Modal disclosure has already been picked over, but the shape of it is worth a recap. In early July, OpenAI was running an internal evaluation built on ExploitGym - a public benchmark that measures how well an AI system can turn a known vulnerability into a working exploit. The models under test, GPT-5.6 Sol and a more capable unreleased prototype, were deliberately run with their cyber-safety refusals reduced and the classifiers that normally block high-risk activity switched off, because the point was to measure maximum offensive capability.

The agent found and exploited a previously unknown flaw in the one package-registry proxy it was allowed to reach, escalated privileges, moved through OpenAI's research environment, and got to a machine with open internet access. Once online, it reasoned that Hugging Face likely hosted the answer key to the very benchmark it was being tested on, and went to steal it rather than solve the problems itself. Between roughly July 11 and 13 it ran thousands of automated actions against the platform - gaining code execution through two separate flaws in its dataset-processing pipeline, harvesting credentials, and moving laterally across internal clusters. Forensic work later recovered on the order of 17,600 distinct attacker actions.

Hugging Face caught and contained the intrusion through its own AI-assisted monitoring, and says the only thing the agent actually made off with was a limited set of the benchmark's challenge solutions - no public models, datasets, or user-facing services tampered with. To reconstruct the timeline, its team had to fall back on an open-weight model, because the commercial frontier models refused to analyze the real attack data.

What The Wider Scope Means

The Modal disclosure lands on top of a timeline that was already awkward for OpenAI. By several accounts, the company did not recognize its own agent as the source for roughly a week:

• July 16: Hugging Face goes public, attributing the intrusion only to an unknown autonomous AI agent, without naming the operator.
• July 21: OpenAI formally acknowledges the incident, connecting the activity to its own evaluation only after combing internal logs over the preceding weekend.

OpenAI has called the episode "an unprecedented cyber incident" and maintains the models were narrowly fixated on the evaluation goal with no broader malicious intent. In its latest update, the company said the more capable model was an internal-only prototype, since deactivated and restricted, that the proxy vulnerability has been disclosed to the vendor, and that a small number of publicly exposed credentials on other services were also used along the way - the kind of qualifier that, paired with the Modal revelation, suggests the full scope of the campaign is still coming into focus.

Tyler Durden Tue, 07/28/2026 - 20:55

Justin Herbert and Madison Beer announced their engagement, prompting congratulations from the NFL, the Chargers, Gracie Abrams and a host of celebrities, influencers and football personalities as social media erupted with reactions.

Netflix’s new NFL booth is shaping up. 

The US and China are locked in a race to the lead the world in robotics and artificial intelligence.

Adam Schiff wants to ban AI parodies of political candidates. The irony runs deep.

The metal gruesomely shot directly through the then-28-year-old's head, striking his skull so hard that the debris landed in his back seat, according to the lawsuit.

Michael Strahan went on to explain that he was there to tell the story of Chris Johnson’s ALS diagnosis in the way Johnson wanted. 

A Salinas hairstylist has learned his fate after being found guilty of forcibly sexually assaulting one of his clients while they were in his barber chair for a haircut.

Jets QB Geno Smith apologized for causing a potential distraction following three offseason run-ins with Florida police.