Skip to main content

Researchers And Army Join Hands to Protect the Military’s AI Systems


As an initiative to provide protection to the military's artificial intelligence systems from cyber-attacks, researchers from Delhi University and the Army have joined hands, as per a recent Army news release. 

As the Army increasingly utilizes AI frameworks to identify dangers, the Army Research Office is investing in more security. This move was a very calculated one in fact as it drew reference from the NYU supported CSAW HackML competition in 2019 where one of the many major goals was to develop such a software that would prevent cyber attackers from hacking into the facial and object recognition software the military uses to further train its AI.

MaryAnne Fields, program manager for the ARO's intelligent systems, said in a statement, "Object recognition is a key component of future intelligent systems, and the Army must safeguard these systems from cyber-attack. This work will lay the foundations for recognizing and mitigating backdoor attacks in which the data used to train the object recognition system is subtly altered to give incorrect answers."


This image demonstrates how an object, like the hat in this series of photos, can be used by a hacker to corrupt data training an AI system in facial and object recognition.
The news release clearly laid accentuation on a very few important facts like, “The hackers could create a trigger, like a hat or flower, to corrupt images being used to train the AI system and the system would then learn incorrect labels and create models that make the wrong predictions of what an image contains.” 

The winners of the HackML competition, Duke University researchers Yukan Yang and Ximing Qiao, created a program that can 'flag and discover potential triggers'. And later added in a news release, "To identify a backdoor trigger, you must essentially find out three unknown variables: which class the trigger was injected into, where the attacker placed the trigger and what the trigger looks like," 

And now the Army will only require a program that can 'neutralize the trigger', however, Qiao said it ought to be "simple:" they'll just need to retrain the AI model to ignore it. 

And lastly, the software's advancement is said to have been financed by a Short-Term Innovative Research that grants researchers up to $60,000 for their nine months of work.



source https://www.ehackingnews.com/2020/01/researchers-and-army-join-hands-to.html
Labels:

Comments

Post a Comment

Popular posts from this blog

Betting and Gambling Websites under Cyberattack from Chinese Hackers

Since last year's summers, Chinese hackers have been targeting South Asian companies that own online gambling and betting websites. The gambling companies in South Asia have confirmed the hacks, whereas rumors of cyberattacks on betting websites have also emerged from Europe, and the Middle East, however, the rumors are yet to confirm, says the reports of cybersecurity group Trend Micro and Talent-Jump. Cybersecurity experts claim that no money was stolen in these hacks against the gambling websites. However, hackers have stolen source codes and databases. The motive of the attack was not a cybercrime, but rather espionage intended attack to gain intelligence. According to the experts, a group named ' DRBControl ' is responsible for the cyberattack. According to the reports of Trend Micro, the hacking techniques used in this particular cyberattack incident is similar to methods done by Emissary Panda and Winnti. All of these hacking groups are from China that has launc
5 comments
Read more

Information security experts have warned Russians about bonus card fraud schemes.

Fraudsters several thousand times tried to illegally take advantage of discount bonuses of Russians in 2019. Some attackers gained access to customers' personal accounts, and then bought the products using bonuses, said Alexey Sizov, head of the anti-fraud department of the Application Security Systems Center at Jet Infosystems. According to him, a fraudster can register a personal account on a card that was issued to another person. The victim will accumulate points without knowing about the existence of his profile, and the attacker will write off bonuses, said Sizov. The expert added that this is mainly done by novice scammers. According to him, loyalty programs are poorly protected, unlike banking operations. He said that they are estimated at 50 billion rubles ($760 milliard) for the 30 largest retailers. Alexey Fedorov, Chairman of the Business Russia Trade Committee, said that in 2019, the number of bonus and discount thefts "increased significantly."
Post a Comment
Read more

Provider Volia reported to the cyber police about the intense cyberattacks on the server

Cable provider Volia appealed to the Cyber Police on the fact of fixing a DDoS attack on the Kharkov servers of the company, which has been ongoing since May 31. "For three days, from May 31 to today, the Volia infrastructure in Kharkov is subjected to cyberattacks. At first, they were carried out only on subscriber subsystems, later they switched to telecommunications infrastructure. As a result, more than 100,000 subscribers experienced problems using the Internet, IPTV, multi-screen platform, and digital TV," said the company. In total, the complete lack of access to Volia's services, according to the provider, lasted 12 minutes on May 31, 45 minutes on June 1. There was also an attack on the website volia.com, but it was managed to neutralize. "DDoS attacks were massive and well-organized. The type of attack is UDP flood and channel capacity overflow with the traffic of more than 200 GB. UDP is a protocol used for online streaming services - streaming, te
Post a Comment
Read more