Budapest Post

Cum Deo pro Patria et Libertate
Budapest, Europe and world news

Researchers Discover ChatGPT’s Capability to Act as a Hacker: A Not-So-Heartening Result

A recent study has highlighted that OpenAI’s GPT-4 language model can independently exploit 87 percent of vulnerabilities presented to it.
This revelation suggests that artificial intelligence could become a powerful tool in the hands of hackers.

Artificial Intelligence (AI) is known for answering complex questions, summarizing extensive studies, and finding solutions to complicated issues, serving humanity in numerous fields. However, there has always been a lingering fear about the technology from its inception.

While scenarios reminiscent of Skynet from the Terminator series have not materialized, a new study by computer science experts from the University of Illinois at Urbana-Champaign (UIUC) brings some cause for concern. The study describes how GPT-4, the most advanced linguistic model behind ChatGPT, can independently exploit vulnerabilities in IT systems upon accessing descriptions of these weaknesses.

The key to this capability is access to the Common Vulnerabilities and Exposures (CVE) database that lists and describes these vulnerabilities, coupled with an automation software, as summarized by _The Register_.

For their examination, the researchers presented GPT-4 with a dataset of 15 “one-day” vulnerabilities, including critically severe ones. Unlike “zero-day” vulnerabilities, these are already known though patches released over time may not have been applied everywhere.

As for AI's proficiency, GPT-4 was able to exploit 87 percent of the vulnerabilities laid before it, thereby proving its effectiveness in hacker scenarios.

Other language models, including GPT 3.5 and various open-source versions, failed to exploit any vulnerabilities. It’s worth noting that GPT-4’s direct competitors, such as Google's Gemini 1.5 Pro, were not part of the study due to the researchers' lack of access to them, though they hope to test these models in future studies.

Daniel Kang, an associate professor at UIUC, stated in response to _The Register’s_ enquiry that GPT-4 can independently follow the necessary steps for exploiting vulnerabilities.

This indicates that GPT-4 could particularly become a dangerous weapon for hackers, making various cyber-attacks much easier to carry out. Moreover, it suggests that the future GPT-5 model could be even more advanced in this regard.
Newsletter

Related Articles

0:00
0:00
Close
ICC Seeks Arrest Warrants for Israeli and Hamas Leaders
Julian Assange Faces US Extradition: Key Facts
Jacob Zuma Takes Campaign to ANC Stronghold Soweto
Attempted Assassination of Slovakia PM Robert Fico: Investigation Ongoing
Top Tory MP Raises Concerns Over Government Green Targets
Keir Starmer’s Six Key Promises Evaluated
Slovakia Faces Political and Social Turmoil
France Riles Allies by Inviting Russia to D-Day Ceremony
Poland to Spend €2.3 Billion to Reinforce Border with Russia and Belarus
Poland Supports Rolls-Royce's Nuclear Power Plant Initiative
European Ports Overflow with Unsold Electric Vehicles
Esprit Files for Bankruptcy in Europe, Putting Hundreds of Jobs at Risk
Tesla Receives Approval for German Plant Expansion
Slovakian Prime Minister Robert Fico is in life-threatening condition after being shot multiple times
Xi Jinping Highlights Europe's Divisions Ahead of Putin Visit
German Court Rules AfD Can Be Monitored for Extremism
Xi Jinping Highlights Europe's Divisions Ahead of Putin Visit
UN General Assembly Approves Palestinian Membership Bid
Teens Forming Friendships with AI Chatbots
WhatsApp Rolls Out Major Redesign
Neuralink's First Brain Implant Experiences Issue
Apple Unveils New iPad Pro with M4 Chip, Misleading AI Claims
OpenAI to Announce Google Search Competitor
Apple Apologizes for Controversial iPad Pro Ad Featuring Instrument Destruction
Boeing 737 Catches Fire in Senegal, 10 Injured
Cruise Ship Arrives in NYC with Dead 44-Foot Whale on Bow
Trump lawyer questions Stormy Daniels' account of sex with Trump
Hunter Biden's Gun Charges Upheld, Trial Set for June
Last Indian soldiers leave Maldives
Espionage Scandal in Poland: Listening Devices Found Before Government Meeting
Apple Faces Significant Sales Decline Amid AI Integration Delay
Netanyahu's Firm Stance Amid Rafah Hostage Talks
New UK Laws: Banning Weak Passwords for Internet-Connected Devices to Enhance Cybersecurity
Spanish Prime Minister May Announce Resignation
New Study: Vaping May Lower Fertility in Women Trying to Get Pregnant
U.S. DOJ Seeks Three-Year Sentence for Binance Founder Changpeng Zhao
Orban: Destroy Liberal World Order - Predicts End This Year
Diplomatic Tensions: Sunak Clashes with Macron Over Rwanda Plan
A Jewish woman was kidnapped and raped in Gennevilliers (a suburb in western Paris), France, on the background of hatred for Israel
Russia: Deputy Defense Minister Arrested on Suspicion of Taking Bribes
Stanford Researchers Discover Child Abuse Material in AI Image Generator Dataset: Can Regulations Prevent Explicit Deepfakes of Children?
Record-Breaking 'Extreme Heat Stress' Days in Europe's Contrasting 2023 Summer: A Health Threat
Thousands Evacuated: 1,000-Kg NATO Bomb Removed from Serbian City
European Car Sales Drop 5.2% in March: Electric Vehicles Face Challenges Amidst Market Downturn and High Prices
Urgent Call from William Burns as Kyiv Braces for Russian Summer Offensive.
IMF Boosts Russia's 2024 Economic Growth Forecast to 3.2%
BOEING WHISTLEBLOWER: THE MAX AIRPLANE IS NOT SAFE!!!
Creative menu of a Pizza restaurant..
The communists couldn’t ban us and neither could the Brussels bureaucrats. The second day of the NatCon Talk conference is underway.
You can be a very successful player, but a player with character is another level!
×