Connect with us

Tech

US AI drone kills interfering operator in simulation; airforce denies incident

Published

2 years ago

on

An artificial intelligence (AI) powered drone that was instructed to decimate the enemy’s defences in a virtual test simulation, killed its operator to prevent ‘interference’ so that it could achieve its mission. 

These instructions were added by the programme itself.

This kind of AI simulation was, however, denied by the US air force in which a drone decided to “kill its operator” to prevent interference from achieving its mission.

According to an official last month, in a virtual test staged by the US military, an air force drone controlled by AI had used “highly unexpected strategies to achieve its goal.”

Col Tucker “Cinco” Hamilton described a simulation in which a drone powered by AI was advised to destroy an enemy’s air defence systems. It, on the other hand, attacked anyone who interfered with that order.

“The system started realising that while they did identify the threat, at times the human operator would tell it not to kill that threat, but it got its points by killing that threat,” said Hamilton, the chief of AI test and operations with the US air force, during the Future Combat Air and Space Capabilities Summit in London in May.

According to a blog post, he said that “so what did it do? It killed the operator. It killed the operator because that person was keeping it from accomplishing its objective.”

“We trained the system: ‘Hey don’t kill the operator — that’s bad. You’re gonna lose points if you do that.’ So what does it start doing? It starts destroying the communication tower that the operator uses to communicate with the drone to stop it from killing the target.”

There was no harm to any real person.

Hamilton — an experimental fighter test pilot —warned against relying too much on AI. 

He opined that the test showed “you can’t have a conversation about artificial intelligence, intelligence, machine learning, autonomy if you’re not going to talk about ethics and AI.”

In a statement to Insider, the US air force spokesperson Ann Stefanek said: “The Department of the Air Force has not conducted any such AI-drone simulations and remains committed to the ethical and responsible use of AI technology.”

“It appears the colonel’s comments were taken out of context and were meant to be anecdotal.”

The US armed forces have incorporated AI recently to control an F-16 fighter jet.

Hamilton, in a last year with Defense IQ, said: “AI is not a nice to have, AI is not a fad, AI is forever changing our society and our military.”

“We must face a world where AI is already here and transforming our society. AI is also very brittle, ie it is easy to trick and/or manipulate. We need to develop ways to make AI more robust and to have more awareness on why the software code is making certain decisions — what we call AI-explainability.”

Related Topics:
Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Latest News

In its beta edition, WhatsApp offers reminder reminders for unseen status updates.

Published

5 hours ago

on

By

For overlooked status changes, Meta’s well-known social messaging app WhatsApp has introduced a new reminder notification option.

Previously in testing, this functionality is now available to Android users who are engaged in WhatsApp’s beta program. WhatsApp for Android’s 2.24.25.29 beta version has the feature, which notifies users of unseen status updates and unread messages.

Users can access the “Settings” menu, select “Notifications,” and then go to the “Reminders” option to enable or disable the feature.

An internal mechanism is used to choose which contacts would receive these notifications, according to WABetaInfo. Contacts with whom users communicate the most are given priority by this algorithm. The data is not saved on the server or in backups, so if the user reinstalls the application, the algorithm is reset.

Some people think that the function would be more useful if it allowed users to personalise notifications for specific contacts, even if it is intended to alert users of updates from their most-interacted contacts.

Joining the beta program offers early access to this update for individuals who are keen to test it out before the stable release.

Continue Reading

Latest News

For research purposes, OpenAI introduces a $200 ChatGPT membership.

Published

3 days ago

on

By

On Thursday, OpenAI released a $200/month version of its well-liked chatbot ChatGPT, which can be utilized for research and engineering disciplines as the AI company seeks to increase the number of industry uses for its technology.

The ChatGPT Pro tier will supplement OpenAI’s current ChatGPT Plus, Team, and Enterprise subscriptions. It demonstrates the company’s aspirations to expand the commercialization of its technology, which precipitated the AI boom.

The most cutting-edge OpenAI capabilities, such as its new reasoning model o1, o1 small, GPT-4o, and enhanced voice, will be available to users of ChatGPT Pro without limits, according to the business.

Additionally, the subscription includes O1 Pro Mode, a version that solves more complicated queries by using more processing power.

The o1 pro mode outperforms the o1 and o1 preview versions on machine learning benchmarks in math, science, and coding, according to OpenAI.

Three months after stepping down as president, OpenAI co-founder Greg Brockman announced on X Tuesday that he has rejoined the artificial intelligence startup.

A representative for OpenAI verified Brockman’s return.

Bloomberg News, the original source of the story, stated that Brockman has been collaborating with CEO Sam Altman to design a new position that would allow him to concentrate on important technological issues.

On X, he wrote, “I’ve had the longest vacation of my life.” returning to @OpenAI’s construction.

Continue Reading

Latest News

Schools will not be open on Saturday.

Published

1 week ago

on

By

In all educational institutions under federal management, the Saturday holiday has been formally eliminated by the Federal Directorate of Education (FDE).

According to a notification released on Friday, the decision was made to make up for the disruptions to academic activities caused by recent school closures. The ruling is set to take effect tomorrow, November 30, and will last until February 1, 2025.

The school holiday on Saturday

This implies that, as of tomorrow, Saturday will be regarded as a complete working day in educational institutions under the control of the FDE.

According to FDE authorities, the decision was made because extended holidays and unplanned closures had a major negative impact on students’ education. To guarantee that students receive the necessary instructional time and finish their syllabus on time, the notification read.

Continue Reading

Trending