Start your day with intelligence. Get The OODA Daily Pulse.

Home > OODA Analysis and Briefs

Analysis

  • Air Security: Not Serious

    A top level security probe is under way after a 12-year-old boy walked on to a plane unchecked with no documents at the height of the terror scare. The boy boarded a plane at Gatwick on Monday despite airport security being on red-alert.   Want more data points that support the idea that – mechanical…

  • Lesson Not Learned

    The director of the Canadian Security Intelligence Service said Tuesday that the spy agency avoids racial profiling because it is “fundamentally stupid” and does not knowingly use information gleaned under torture offshore because the practice is “morally repugnant.” James Judd told a gathering of Canadian judges on Tuesday that he is “acutely aware” of complaints…

  • The New Collection / Production Management

    In my embargoed contribution to a pending book on reform, I argue that going the networked/wiki-ed/blogged route is both a nice way to cut out a lot of the overhead associated with the intel process. As it works now it is very hierarchical, with gatekeepers at every turn and operating at a pace that makes tortoises look…

  • Let the Flailing Begin

    So controversial is the forthcoming National Intelligence Estimate on Iraq, say officials and observers, that even the terms of reference are likely to be the subject of intense, but highly secret, discussion and debate. Somewhere in a conference room in Northern Virginia, roughly two-dozen seniors and their backups are sitting around a table and bickering…

  • Here Comes the Judge

    The plot has also revealed the indispensability of good counterterrorism intelligence. A defense against terrorists, as against other enemies of the nation, must be multilayered to have a reasonable chance of being effective. One of the outer defenses is intelligence, designed to detect plots in advance so that they can be thwarted. One of the…

  • As they bloody well should

    The [UK] Government is discussing with airport operators plans to introduce a screening system that allows security staff to focus on those passengers who pose the greatest risk. The passenger-profiling technique involves selecting people who are behaving suspiciously, have an unusual travel pattern or, most controversially, have a certain ethnic or religious background. The system…

  • No connection . . .

    . . . execpt for the tactic. Incredible: The FBI said Monday it had no information to indicate that the three Texas men arrested in Michigan with about 1,000 cell phones in their van had any direct connection to known terrorist groups. Also, a prosecutor in a separate Ohio case said he can¹t prove a…

  • I don’t buy it

    This is what passes for CT analysis at Time. Acceptance that this is less an AQ issue than it is a terrorism issue is a nice change of pace from the class that loves to poo-poo any disrupted plot that doesn’t reveal the perps had wallets full of AQ business cards. Note however the quick…

  • The Difference Between LE and DI

    As a friend from one of the Commonwealth nations points out, “Nice job by the domestic intelligence service,” which is his subtle but well-meaning dig at both our FBI and larger intelligence apparatus. Domestic tip-off; foreign intelligence supplement; and allied intel cooperation all operationalized for the door-kickers inside of a year. Were this a US…

  • Clarity, Quick Hits & A Suggestion

    As I alluded to earlier, it is a little hard to accurately assess things if you’re not prepared to use accurate if un-PC language. US and UK media might have an aversion to mentioning the Muslim connection, but the Italian media apparently does not. Ethnic and religious descriptions are hardly immaterial in situations like this,…

  • More Terror Plot Short Takes

    Bloody day job . . . Surveillance, electronic or otherwise, works:  By late 2005, the probe had expanded to involve several hundred investigators on three continents. They kept dozens of suspects under close surveillance for months, even as some of the plotters traveled between Britain and Pakistan to raise money, find recruits and refine their…

  • Go Big Blue

    As a guy who knows a thing or two about warning systems, I watched again with dismay the misuse and abuse of the homeland security advisory system today. Granted, it isn’t really a “warning” system per se, but the general concept is the same: give people a quick way to assess the threat. The problem…

  • Curious

    This by Evan Kohlmann on the CT Blog yesterday 8/9: Participants on key Arabic-language Al-Qaida chat forums on the Internet have announced their intent to launch collective cyberterrorist attacks tomorrow, August 10, aimed at interrupting or denying service to other pro-Israeli websites–namely the “Internet Haganah” cyberterror watchdog site run by Aaron Weisburd. […] . .…

  • UK Airline Terror Plot

    What a day to sleep in. Some quick hits before I descend into the seventh level of teleconference hell: In the words of Secretary Chertoff this latest plot is “suggestive” of al-Qaida because it is essentially a dusting off of the plans for Operation Bojinka; funded by al-Qaida and put together by two guys whose…

  • Everything you thought you knew . . .

    . . . about everything covered in the MSM – especially war – is suspect. Some of you didn’t need confirmation, but since I have a nice side-bar going with former info ops colleagues a post seemed appropriate. Army of Analysts Experts people, Army of Experts . . .

Briefs

  • How MIT’s Liquid Neural Networks can solve AI problems from robotics to self-driving cars

    In the current artificial intelligence (AI) landscape, the buzz around large language models (LLMs) has led to a race toward creating increasingly larger neural networks. However, not every application can support the computational and memory demands of very large deep learning models. The constraints of these environments have led to some interesting research directions. Liquid neural…

  • UK calls artificial intelligence a “chronic risk” to its national security

    Artificial intelligence (AI) has been officially classed as a security threat to the UK for the first time following the publication of the National Risk Register (NRR) 2023. The extensive document details the various threats that could have a significant impact on the UK’s safety, security, or critical systems at a national level. The latest…

  • Worldcoin cryptocurrency project suspended in Kenya

    The new cryptocurrency and digital identity project, Worldcoin, is facing a rough start in Kenya, with the government halting all local activity associated with it. Kenya’s minister of internal security took to Facebook on Aug. 2 to announce that the country has suspended the activities of Worldcoin until relevant public agencies had certified the absence…

  • The AI rules that US policymakers are considering, explained

    AI is getting seriously good. And the federal government is finally getting serious about AI. The White House announced a suite of artificial intelligence policies in May. More recently, they brokered a number of voluntary safety commitments from leading AI companies in July. That included commitments to both internal and third-party testing of AI products to…

  • Artificial intelligence flies XQ-58A Valkyrie drone

    Artificial intelligence software successfully flew an XQ-58A Valkyrie drone, the Air Force Research Laboratory announced Aug. 2. The U.S. lab led the three-hour sortie on July 25 with test units at the Eglin Test and Training Complex in Florida. The flight followed two years of work and a partnership with Skyborg Vanguard, a team made…

  • Lake Victoria boat accident kills at least 20 in Uganda; several missing

    At least 20 people have died and five are missing after an overloaded boat capsized in the waters of Lake Victoria in Uganda. The boat was carrying charcoal, fresh foods, and fish when it capsized due to overloading and bad weather. Nine passengers were rescued, and efforts to locate the missing individuals are underway. Lake…

  • Niger power blackouts blamed on coup sanctions

    Major cities in Niger are experiencing rolling blackouts following the recent coup in the country, which has led to Nigeria cutting electricity supplies to its northern neighbor. The power shortages are affecting cities such as Niamey, Maradi, and Zinder, with residents experiencing intermittent power for about an hour at a time before it is switched…

  • Ukraine war: Drones target Odesa grain stores near Romania border

    A Russian drone strike has targeted Ukrainian port facilities in Izmail on the River Danube, damaging a grain warehouse and an elevator used for loading grain. Approximately 40,000 tonnes of grain destined for various countries, including African nations, China, and Israel, were damaged in the attack. Russia began targeting Ukraine’s ports after abandoning a UN…

  • Brazil police raids leave at least 43 people dead

    Police raids targeting drug gangs in three Brazilian states have resulted in at least 43 deaths. In Rio de Janeiro, a shootout in the Complexo da Penha area led to the deaths of at least 10 people. In São Paulo state, 14 people died in clashes during a five-day police raid known as Operation Shield.…

  • Google AMP Abused in Phishing Attacks Aimed at Enterprise Users

    Threat actors have been exploiting Google Accelerated Mobile Pages (AMP) in phishing campaigns as a new tactic to evade detection, according to cybersecurity firm Cofense. Google AMP is an open source HTML framework designed to improve mobile page performance. Attackers have been abusing a Google AMP feature that allows newly created pages to be hosted…

  • Firefox 116 Patches High-Severity Vulnerabilities

    Mozilla has released Firefox 116, Firefox ESR 115.1, and Firefox ESR 102.14, which include patches for multiple high-severity vulnerabilities. The update addresses a total of 14 CVEs (Common Vulnerabilities and Exposures), with nine of them rated as high severity. Some of the vulnerabilities include cross-origin restrictions bypass, use of an incorrect value during WASM compilation,…

  • Cyble Raises $24 Million for AI-Powered Threat Intelligence Platform

    Cybersecurity firm Cyble has successfully raised $24 million in a Series B funding round, bringing its total funding to over $38 million. Co-led by Blackbird Ventures and King River Capital, the funding will be used to bolster the Atlanta-based company’s AI-powered security solutions and support its global expansion. Cyble specializes in gathering intelligence from various…

  • New hVNC macOS Malware Advertised on Hacker Forum

    Israeli cybersecurity company Guardz has issued a warning about a new form of Hidden Virtual Network Computing (hVNC) malware specifically targeting macOS devices. This malware, being advertised on a prominent cybercrime forum, allows threat actors to gain control of remote systems without the user’s knowledge. The malware is offered at a price of $60,000 and…

  • Google’s Jigsaw was trying to fight toxic speech with AI. Then the AI started talking

    Human abuse was hard enough. In 2017, a team at Google released Perspective API, an AI-based tool designed to help flag the toxic speech that pushes people out of online spaces, pushes them to violence, or worse. Platforms like YouTube and Facebook were already building their own AI classifiers to battle all sorts of hate speech,…

  • Voyager and Airbus will operate Starlab private space station via ‘transatlantic joint venture’

    Voyager Space and Airbus Defence and Space are expanding their existing relationship via a new joint venture focused on the design, build and operation of the Starlab commercial space station. The two companies announced that they would be partnering on Voyager’s Starlab space station in January, saying that Airbus would provide “technical design support and expertise,”…

  • Nvidia AI Image Generator Fits on a Floppy Disk and Takes 4 Minutes to Train

    In the rapidly evolving landscape of AI art creation tools, Nvidia researchers have introduced an innovative new text-to-image personalization method called Perfusion. But it’s not a million-dollar super heavyweight model like its competitors. With a size of just 100KB and a 4-minute training time, Perfusion allows significant creative flexibility in portraying personalized concepts while maintaining…

  • New DeepMind AI model can control robotic actions it was never trained to do

    Google DeepMind has a new AI model that can direct robotic tasks it was never trained to perform. Named RT-2, the model learns from web and robotics data. It then turns this information into simple instructions for machines. In tests, the model was asked to take actions never seen in the robotic data, such as placing oranges…

  • Meta’s AI music generator could be the new synthesizer — or just muzak

    Meta released a new open-source AI code called AudioCraft, which lets users create music and sounds entirely through generative AI. It consists of three AI models, all tackling different areas of sound generation. MusicGen takes text inputs to generate music. This model was trained on “20,000 hours of music owned by Meta or licensed specifically…

  • AI reduces data breach lifecycles and costs

    The cybersecurity tools you implement can make a difference in the financial future of your business. According to the 2023 IBM Cost of a Data Breach report, organizations using security AI and automation incurred fewer data breach costs compared to businesses not using AI-based cybersecurity tools. The report found that the more an organization uses the…

  • Micron Just Changed the Game in Artificial-Intelligence Memory

    Micron just unveiled specs for its new high-bandwidth memory for AI, which appears to one-up the fastest HBM on the market. Leading memory producer Micron Technology and its peers are in one of the worst memory downturns in history, but the hope is that memory-hungry artificial-intelligence (AI) applications will help the industry dig itself out. Micron…

  • Nasa Voyager 2: Space agency accidentally loses contact with pioneering space probe

    Nasa has “inadvertently” lost contact with Voyager 2, its pioneering spacecraft. The space agency said that it had sent a series of planned commands to the probe on 21 July. But those moves left Voyager 2 pointing in the wrong direction: its antenna is now pointing two degrees away from Earth, and so it has…

  • Artificial intelligence threats in identity management

    The 2023 Identity Security Threat Landscape Report from CyberArk identified some valuable insights. 2,300 security professionals surveyed responded with some sobering figures: 68% are concerned about insider threats from employee layoffs and churn 99% expect some type of identity compromise driven by financial cutbacks, geopolitical factors, cloud applications and hybrid work environments 74% are concerned…

  • July Crypto Hack Spree Threatens Recovery of DeFi Sector

    Crypto hacks and exploiters had a ball in July with around 50 attacks resulting in hundreds of millions in losses. An equally bumper August for cyber criminals could hamper any recovery for the decentralized finance (DeFi) sector. According to blockchain security firm PeckShield, there were more than 48 major hacks in July, leading to around…

  • Cybercriminals train AI chatbots for phishing, malware attacks

    In the wake of WormGPT, a ChatGPT clone trained on malware-focused data, a new generative artificial intelligence hacking tool called FraudGPT has emerged, and at least another one is under development that is allegedly based on Google’s AI experiment, Bard. Both AI-powered bots are the work of the same individual, who appears to be deep…

  • Sweden rules out sweeping changes to free speech law amid Quran burnings

    Sweden’s government has stated that it has no plans to make significant changes to its freedom of speech laws. However, it will explore measures that would allow police to intervene and prevent the burning of holy books in public if there is a clear threat to national security. The recent incidents of Quran burning in…

  • Voyager 2: Nasa loses contact with probe after sending wrong command

    NASA has lost contact with its Voyager 2 probe after sending it the wrong command, causing the spacecraft to tilt its antenna away from Earth. The probe, which has been exploring the universe since 1977, is currently more than 12.3 billion miles away from Earth in interstellar space. NASA hopes that communication will resume when…

  • Xi Jinping replaces leaders of China’s elite nuclear force

    China has replaced two leaders of its elite unit managing its nuclear arsenal, leading to speculation of a purge. General Li Yuchao and his deputy, who headed the People’s Liberation Army’s (PLA) Rocket Force unit, had been “disappeared” for months. The replacements, former deputy navy chief Wang Houbin and party central committee member Xu Xisheng,…

  • Niger coup: France to evacuate citizens after embassy attack

    Following the recent coup in Niger, anti-French sentiment has led to attacks on the French embassy and protests outside the embassy in Niamey. As a result, France has decided to evacuate its citizens from Niger. Germany and Italy are also organizing flights for their nationals to leave the country. The coup has caused demonstrations against…

  • Ransomware Attacks on Industrial Organizations Doubled in Past Year: Report

    Data from industrial cybersecurity firm Dragos shows that the number of ransomware attacks targeting industrial organizations and infrastructure has doubled since the second quarter of 2022. In the second quarter of 2023, Dragos observed 253 ransomware incidents, an 18% increase from the previous quarter. The surge in attacks has been attributed to ransomware revenue plunging…

  • Silk Security Emerges from Stealth With $12.5 Million Seed Funding

    Silk Security, a New York-based startup, has raised $12.5 million in a seed funding round led by Insight Partners and Hetz Ventures. The company aims to break down the silos between security and development with its integrated “find and fix” platform. Silk consolidates alerts from various detection systems, performs automatic triaging based on threats, and…

  • SpecterOps Updates BloodHound Active Directory Mapping Tool

    BloodHound, developed by SpecterOps, is a graph-based map of Active Directory (AD) relationships used by external pentesters and internal users to discover potential lateral movement attack paths. Active Directory is a highly prized target for attackers due to its ubiquity and complexity. BloodHound helps find AD misconfigurations that could be exploited by attackers. SpecterOps has…

  • Socket Scores $20M as Investors Bet on Software Supply Chain Security Startups

    Socket, a startup focused on software supply chain security, has raised $20 million in a recent funding round led by Andreessen Horowitz (a16z) and Abstract Ventures. The company, founded by open source developer Feross Aboukhadijeh, uses “content-based analysis” to detect suspicious activity and behavior in software dependencies. Its technology aims to prevent compromised packages from…

  • The Robots We Were Afraid of Are Already Here

    Digit drew a crowd, even here, in a convention center full of robot aficionados. A humanoid warehouse worker, Digit walked upright on goatlike legs and grabbed bins off a shelf with muscular arms made from aerospace-grade aluminum. It then placed the boxes on an assembly line and walked back to the shelf to search for more.…

  • Cryptography may offer a solution to the massive AI-labeling problem

    The White House wants big AI companies to disclose when content has been created using artificial intelligence, and very soon the EU will require some tech platforms to label their AI-generated images, audio, and video with “prominent markings” disclosing their synthetic origins. There’s a big problem, though: identifying material that was created by artificial intelligence is…

  • The Rise of Artificial Intelligence and its Potential Impact

    Artificial intelligence (AI) is a creation of humans, designed to enhance the capabilities of robots, computers, and other specialized machines. While some have reservations about this man-made intelligence, it is important to recognize the possibilities it offers in our increasingly complex world. One intriguing aspect of AI is its ability to communicate with humans. These simulated…

  • Companies double down on AI in June-quarter analyst calls

    It’s a high bar, but companies reporting second-quarter earnings in recent weeks have talked up artificial intelligence even more than in the previous quarter. S&P 500 companies that led in discussion of AI during quarterly conference calls with analysts earlier this year have outdone themselves in their latest quarterly calls. Following Intel’s report late on Thursday, executives…

  • Curve Finance pools exploited by over $47M due to reentrancy vulnerability

    Several stable pools on Curve Finance using Vyper were exploited on July 30, with losses reaching over $47 million. According to Vyper, its 0.2.15, 0.2.16 and 0.3.0 versions are vulnerable to malfunctioning reentrancy locks. “The investigation is ongoing but any project relying on these versions should immediately reach out to us,” Vyper wrote on X.…

  • CISA Analyzes Malware Used in Barracuda ESG Attacks

    The US Cybersecurity and Infrastructure Security Agency (CISA) has published analysis reports on three malware families used in an attack that exploited a recent remote command injection vulnerability (CVE-2023-2868) in Barracuda Email Security Gateway (ESG). The vulnerability was a zero-day and affected versions 5.1.3.001 to 9.2.0.006 of the appliance. A Chinese state-sponsored cyberespionage group, UNC4841,…

  • Second Ivanti EPMM Zero-Day Vulnerability Exploited in Targeted Attacks

    Ivanti, a software company, has issued a warning to its customers about a second zero-day vulnerability in its Endpoint Manager Mobile (EPMM) product. The first zero-day, CVE-2023-35078, was exploited in a cyberattack targeting Norwegian government ministries. The newly discovered vulnerability, CVE-2023-35081, allows an authenticated attacker with administrator privileges to remotely write arbitrary files to the…

  • Reddit Taps Fredrick ‘Flee’ Lee for CISO Job

    Reddit has appointed a new Chief Information Security Officer (CISO), Flee Lee, to oversee its privacy and security functions. Lee brings over 20 years of experience in global information security and privacy efforts, previously holding leadership positions at companies like Gusto and Square. As Reddit prepares for an IPO after raising substantial funding, Lee’s expertise…

  • Apple Lists APIs That Developers Can Only Use for Good Reason

    Apple is taking steps to boost user privacy by requiring app developers to declare the reasons for using certain APIs that could potentially be misused for data collection through fingerprinting. The targeted APIs include those used for accessing file timestamps, system boot time, available disk space, active keyboards, and user defaults. Developers will need to…

  • Situation in Niger is ‘cause for serious concern’, says Kremlin

    Russia has said the situation in Niger is cause for serious concern on Monday. The country’s president was ousted in a coup that was condemned largely internationally, except by Wagner mercenary boss Yevgeny Prigozhin, who has extensive interests in Africa. Russian spokesperson Dmitry Peskov called for all sides in Niger to show restraint and return…

  • Pakistan holds funerals as death toll from rally bombing rises to 54

    The death toll from a suicide bombing that targeted an election rally in Pakistan has risen to 54. The election rally was for a pro-Taliban Muslim leader. The government has promised to hunt down those behind the attack. No one immediately claimed responsibility for the attack that occurred on Sunday, which also wounded almost 200…

  • Intel CEO: ‘We’re going to build AI into every platform we build’

    Intel CEO Pat Gelsinger was very bullish on AI during the company’s Q2 2023 earnings call — telling investors that Intel plans to “build AI into every product that we build.” Later this year, Intel will ship Meteor Lake, its first consumer chip with a built-in neural processor for machine learning tasks. (AMD recently did the…

  • The US government is taking a serious step toward space-based nuclear propulsion

    Four years from now, if all goes well, a nuclear-powered rocket engine will launch into space for the first time. The rocket itself will be conventional, but the payload boosted into orbit will be a different matter. NASA announced Wednesday that it is partnering with the US Department of Defense to launch a nuclear-powered rocket engine…

  • Aided by A.I. Language Models, Google’s Robots Are Getting Smart

    A one-armed robot stood in front of a table. On the table sat three plastic figurines: a lion, a whale and a dinosaur. An engineer gave the robot an instruction: “Pick up the extinct animal.” The robot whirred for a moment, then its arm extended and its claw opened and descended. It grabbed the dinosaur.…

  • US Senate approves national defense bill that also targets crypto mixers

    The United States Senate passed the 2024 National Defense Authorization Act (NDAA) worth $886 billion on July 27. The bill includes a provision targeting crypto mixers, anonymity-enhancing coins and institutions engaging in crypto trading. The NDAA is a bill that helps authorize how the country’s defense department can utilize federal funding. Within the bill, a…

  • CoinsPaid Blames North Korean Hackers for $37 Million Cryptocurrency Heist

    CoinsPaid, a cryptocurrency payments firm based in Estonia, has reported a cyberattack resulting in the theft of approximately $37.3 million in cryptocurrency. The attack, which occurred on July 22, involved elements of social engineering, bribery attempts, and attacks on internet-accessible applications. CoinsPaid identified the vulnerability and addressed it before more funds could be stolen, and…

  • Zimbra Patches Exploited Zero-Day Vulnerability

    Zimbra, a popular email and collaboration suite, has released patches for a cross-site scripting (XSS) vulnerability, tracked as CVE-2023-37580, that has been exploited in malicious attacks. The vulnerability was disclosed earlier this month, and in-the-wild exploitation had been observed. Zimbra has now issued software updates for its Collaboration Suite versions 8.8.15, 9.0.0, and 10.0.x, with…

  • Industry Reactions to New SEC Cyber Incident Disclosure Rules: Feedback Friday

    The US Securities and Exchange Commission (SEC) has adopted new cybersecurity incident disclosure rules for public companies, requiring them to disclose security breaches with material impact within four business days. While some see this as a positive step towards transparency and accountability, others are concerned that it could help cybercriminals by providing information they could…

  • Exploitation of Recent Citrix ShareFile RCE Vulnerability Begins

    Threat intelligence company Greynoise has reported the first attempts to exploit a critical remote code execution (RCE) vulnerability in Citrix ShareFile, a popular cloud-based file-sharing and collaboration solution. The vulnerability, tracked as CVE-2023-24489, allows unauthenticated file upload leading to RCE. Cybersecurity firm Assetnote, which identified and reported the bug, estimates there are between 1,000 and…

  • North Korea’s Kim displays nuclear-capable missiles, drones at parade

    North Korean leader Kim Jong Un has displayed nuclear-capable missiles and attack drone in a large military parade in Pyongyang. The parade was overseen by Kim and marked “Victory Day,” North Korea’s name for the end of hostilities in the Korean War in 1950-53. Kim was accompanied by visiting delegations from China and Russia at…

  • Deadly bomb blast near Damascus Shia shrine ahead of Ashura

    At least six people have been killed and over 20 wounded in a bomb explosion near a Shia Muslim shrine south of the capital city of Syria, Damascus. The explosion occurred one day ahead of Ashura, a day of mourning observed by Shia Muslims. The blast was near Syria’s most visited Shia pilgrimage site and…

  • Ruling party HQ attacked after President Bazoum ousted in Niger coup

    Supporters of a coup in Niger attacked the headquarters of the ousted president’s party on Thursday. The headquarters were set on fire and stoned, while cars were also burned outside. The small group of arsonists broke away from a larger show of support for the leaders of the coup that was taking place outside parliament.…

  • Meta sinks $3.7B in metaverse money pit as CEO eyes Threads user retention

    Meta’s metaverse-related losses topped $3.74 billion over the second quarter with the Big Tech player spending $7.7 billion on its virtual reality business so far in 2023. Its second-quarter 2023 results released on July 26 saw Meta report an 11% revenue gain compared to the same quarter last year, totaling $31.9 billion. Its metaverse-focused Reality…

  • Amazon has drawn thousands to try its AI service competing with Microsoft, Google

    Amazon.com’s cloud division has drawn thousands of customers to try out its service vying with Microsoft and Google in a key area of artificial intelligence, an executive told Reuters. The company also announced new AI tools on Wednesday, including a program for building more conversational customer-service agents, technology access from the startup Cohere and a healthcare…

  • NVIDIA DGX Cloud AI Supercomputing Brings AI Training as-a-Service

    NVIDIA’s DGX Cloud infrastructure, which lets organizations lease space on supercomputing hardware suitable for training generative AI models, is now generally available. First announced in March, the $36,999 per instance per month service is in competition with NVIDIA’s own $200,000 DGX server. It runs on Oracle Cloud infrastructure and on NVIDIA hardware located in the…

  • Researchers Poke Holes in Safety Controls of ChatGPT and Other Chatbots

    When artificial intelligence companies build online chatbots, like ChatGPT, Claude and Google Bard, they spend months adding guardrails that are supposed to prevent their systems from generating hate speech, disinformation and other toxic material. Now there is a way to easily poke holes in those safety systems. In a report released on Thursday, researchers at Carnegie Mellon…

  • Lazarus hackers linked to $60 million Alphapo cryptocurrency heist

    Blockchain analysts blame the North Korean Lazarus hacking group for a recent attack on payment processing platform Alphapo where the attackers stole almost $60 million in crypto. Alphapo is a centralized crypto payment provider for gambling sites, e-commerce subscription services, and other online platforms, which was attacked on Sunday, July 23rd, with the initial stolen…

  • Australia awards South Korea’s Hanwha $4.7 billion defense contract

    Hanwha Aerospace is South Korea’s largest defense contractor. On Thursday, the South Korean company beat Germany’s Rheinmetall for an Australian contract worth $4.74 billion. The contract for 129 infantry fighting vehicles (IFVs) is one of the largest in Australian Army history. Hanwha will build the vehicles at their facilities in Australia’s Victoria state. The new…