JPMorgan Sues Customers Over 'Infinite Money Glitch'
JPMorgan Chase has begun suing customers who allegedly stole thousands of dollars from ATMs by taking advantage of a technical glitch that allowed them to withdraw funds before a check bounced. The bank filed lawsuits in at least three federal courts, targeting individuals who withdrew the highest amounts in the 'infinite money glitch' that went viral on social media. The lawsuits seek the return of the stolen funds with interest, overdraft fees, and in some cases, punitive damages. This event underscores the risks of social media amplifying vulnerabilities discovered at financial institutions and the ongoing threat of fraud in the digital age.
Source: https://www.cnbc.com/2024/10/28/jpmorgan-suing-customers-over-infinite-money-glitch.html
Robinhood's Expansion into Election Gambling Raises Concerns
The popular trading app Robinhood has announced the introduction of a new feature allowing users to gamble on the outcome of the upcoming presidential election, framing it as 'democratizing access to events as they unfold.' This expansion into election gambling may signal a broader trend of introducing speculative contracts for users to trade on.
Source: https://www.theverge.com/2024/10/28/24281883/robinhood-presidential-betting
New Startup Turns Carbon Dioxide into Liquid Fuel
SpiralWave, a startup, has developed pulsing plasma towers that can capture carbon dioxide from the atmosphere or a smokestack and transform it into methanol, a liquid fuel. The process is highly efficient, with 75-90% of the system's electrical energy being transformed into chemical energy stored in the form of methanol, making it a promising technology for both carbon removal and fuel production.
Study Finds that Vision-Language Models Struggle with Human-Like Pattern Recognition
A recent study has revealed limitations in the reasoning capabilities of today’s most advanced AI vision-language models, particularly in their struggle to solve problems that require high-level human-like abstract reasoning. While these findings underscore the limitations of current AI, they also point toward opportunities for future innovation and the need for more advanced tests to measure true AI comprehension. The study encourages AI researchers to think beyond existing benchmarks and aim for advancements that bring machines closer to human-level perception and reasoning, ultimately emphasizing the complexity of human cognition and the road AI still has to travel to replicate it.
Source: https://thedebrief.org/29289-2-vision-language-models/
UnitedHealth Data Breach Results in Over 100 Million People's Information Leaked
United Health Group has confirmed a ransomware attack earlier this year that affected the private data of over 100 million people, making it the largest known healthcare data breach. The hackers gained access by using stolen credentials for a Citrix remote access service that lacked multifactor authentication, and United Health paid the group a $22 million ransom. However, another operation threatened to continue leaking the data in an attempt to secure a second ransom payment.
Source: https://www.theverge.com/2024/10/25/24279288/unitedhealth-change-breach-100-million-leak
AI Chatbots Vulnerable to 'Deceptive Delight' Jailbreak Method, Researchers Warn
Researchers have discovered a new AI jailbreak method, named 'Deceptive Delight,' that tricks chatbots into returning restricted content by mixing malicious and benign queries together. This method has shown a 65% success rate against eight different large language models. This discovery highlights the vulnerability of AI chatbots and the need for robust security measures to protect against prompt-injection attacks.
Comments
Post a Comment