The competition , which was in public announced Wednesday at 2023 ’s Blackhat USA cybersecurity conference , will be lock by DARPA , theDefense Department ’s research and development authority , and will hand out nearly $ 20 million in prizes to top - performing contest participant , with the winner getting a sum of $ 4 million in loot money .
In a call with reporters on Tuesday , White House functionary expressed hope that the Modern program would be a first footstep towards create previously unheard - of cyber defenses . Anne Neuberger , deputy national security consultant for Joe Biden ’s White House , said that she felt the challenge would help America “ stay in front in the airstream against our opposer ’ cyber offensive capabilities . ”
Defending critical infrastructure…with a little help from Silicon Valley
The new AI cyber challenge ( abridge “ AIxCC ” ) will have a number of dissimilar phase . Interested would - be competitors can now submit their proposals to the Small Business Innovation Research platform for valuation and , finally , selected teams will take part in a 2024 “ qualify effect . ” During that event , the top 20 squad will be invite to a semifinal competition at that yr ’s DEF CON , another tumid cybersecurity conference , where the field will be further whittle down .
Interestingly enough , a identification number of big AI - focused companies have partner with the governance to assist with the contest . Google , Microsoft , OpenAI ( which is itselfentwined with Microsoft ) , and Anthropic ( which has beenheavily fundedby Google ) will be “ lend their expertise and establish their cutting - bound technology available ” to contest participant , said the White House in a crush release Wednesday . The company have offered to open up their software program systems to repugn participants , who can use the computer software in the crafting of their new security solutions . These same companies also recently take part ina White House summitwhere participants concur to a vague set of diligence “ commitments ” plan to stave off the destructive capabilities that AI poses . At that tip , one of the fundamental areas of discussion was also the Cartesian product between AI and cybersecurity .
To secure the top billet in DARPA ’s new competition , participant will have to develop security solutions that do some seriously novel poppycock . “ To bring home the bacon first - situation , and a top dirty money of $ 4 million , finalists must build a system that can quickly defend critical infrastructure code from attack , ” said Perri Adams , programme manager for DARPA ’s Information Innovation Office , during a Zoom call with newsperson Tuesday . In other words : the authorities desire software that is able of place and mitigating peril by itself .
Anne Neuberger, deputy national security advisorPhoto: Drew Angerer (Getty Images)
The focal point of this project ( critical substructure code ) is an important one . In cybersecurity , the terminus “ critical substructure ” refers to life-sustaining industrial systems — thing like office grids , oil pipelines , water supplies , and other crucial digitally - connected physical systems . In late geezerhood , cyberattacks on such systems have become more and more routine . Most notably , the 2021 ransomwareattack on Colonial Pipeline , which pass off during Biden ’s first yr in office , threatened to edit off lively get-up-and-go hang throughout large part of the southeast United States and was look at awakeup callfor those in the industrial security sphere .
Automating cybersecurity
It ’s comfortable to see this competition as an endeavour by the Biden government to kill two wench with one stone . While the White House has previouslymade it knownthat it need to beef up critical infrastructure security , there ’s also beena lot of chatterabout what AI can do to automate the cybersecurity field . If successful , this contest could fulfill both of those goals at once .
Technically speak , the new program anticipate to automatize the cognitive process ofbug search , which has long been an weak nontextual matter . properly now , the internet is positivelyswimming with security holesand the only seawall against those holes are so - called “ glitch hunting watch , ” security - minded nerds that search for and report exposure to impact companies ( usually for a small fee ) . The future that the new DARPA contest envisions is one in which automated system of rules scan for , uncover , and “ mess ” vulnerabilities with speedy velocity , hopefully stick out badly - intentioned cybercriminals or land - state threat actors that wish to tap the bugs to bring havoc .
That said , it ’s unclear how some of the stillless - than - perfect elementsof existing AI system of rules may affect the human body of the result that are develop . stilted intelligence service is still in its early childhood , really , and it has n’t been made alone percipient yet how competitor player will in reality be using the AI systems provided by some of Silicon Valley ’s heaviest hitter . In unforesightful : how this all shake off out is still unidentified but , suffice it to say , competition members will be boldly plump where infosec has never gone before .
BidenComputer securityCyberattackData protectionMicrosoftOpenAI
Daily Newsletter
Get the honorable technical school , science , and acculturation word in your inbox day by day .
word from the time to come , delivered to your present .
You May Also Like
