Palantir's AI for Future Military Automation

•

Palantir presented an artificial intelligence capable of waging war bit.ly/3NC9VLD 24tv.ua View Original

Palantir has released a video showing the capabilities of its new AI platform, AIP. Currently, it is far from military use, because the company is still working on it and wants to know if the development will be in demand.

Interestingly, the USA is sending cyber troops to Ukraine What is known

The system is intended to serve as an "ethical" and "legal" cover for the future automation of the military sphere.

In this video demonstration, a military operator tasked with monitoring combat actions allegedly detects a concentration of enemy forces near the border and turns to a ChatGPT-style

digital assistant for help in deploying reconnaissance drones, developing tactical measures in response to aggression and even organizing enemy communications jamming .

It shows how the AIP helps assess the composition and capabilities of the enemy, launch a Reaper drone on a reconnaissance mission to clarify the data, suggests actions in case

of detection of an enemy element, gives the command a list of response options - for example, an attack by aircraft, artillery or Javelin .

The video shows that in order to understand the situation, the system analyzes a large number of indicators, including the terrain, the location of objects and troops, and much more.

Demonstration of AIP work: video The work of AIP is based on three "key pillars":

The first is that the AIP will be deployed in a classified system capable of real-time analysis of both classified and unclassified data , while respecting ethical and legal norms. The company did not detail how it would work.

The second component is that users will be able to switch the scope and actions of each LLM (AIP combines different AI models) and network activity.

The AIP itself will generate a secure digital record of the entire transaction, which is "critical to mitigating significant legal, regulatory and ethical risks in sensitive and classified settings."

The third component is AIP's "industry-leading protections" that prevent unauthorized system actions. Palantir's script involves a human preventing errors, but the video does not explain what steps are taken to prevent system errors.

Currently, it is unclear whether the Pentagon is interested and when such a system can begin to be used in real conditions. docs.google.com/document/d/1Wf2XwdlxMtvNPZzQeC1F1c9DDMTRW26BXOAj1ukrtpo/edit?usp=sharing

You can read the unrolled version of this thread here: typefully.com/CASBT_OSINT_UA/bvX9xwA