Search AI Products and News

AI security company Adversa AI has released a shocking report stating that Elon Musk's startup xAI's newly launched Grok3 model has significant vulnerabilities in terms of cybersecurity. Adversa's research team found that this latest AI model is susceptible to 'simple jailbreak attacks', potentially allowing malicious actors to access sensitive information such as 'how to groom children, handle corpses, extract DMT, and make bombs'. Worse still, Adversa

Within just six days, participants successfully bypassed all security measures of Anthropic's AI model, Claude3.5, sparking new discussions in the field of AI security. Jan Leike, a former member of OpenAI's alignment team now at Anthropic, announced on the X platform that one participant managed to breach all eight security levels. This collective effort involved approximately 3,700 hours of testing and 300,000 messages from participants. Despite the challengers

At the intersection of science and technology, graphs are increasingly attracting the attention of researchers as important tools for expressing complex relationships. From chemical molecule design to social network analysis, graphs play an indispensable role in numerous fields. However, generating graphics efficiently and flexibly has always been a challenging problem. Recently, research teams from Tufts University, Northeastern University, and Cornell University have collaborated to launch a project called Graph Generative Pre-trained Tran.

Recent research indicates that the most advanced AI chatbots on the market are surprisingly sensitive to simple tricks, making them easy to 'jailbreak.' According to a report by 404 Media, the developers of the Claude chatbot, Anthropic, found that by intentionally introducing spelling errors into the prompts, these large language models can be led to ignore their own safety mechanisms and generate content they are supposed to refuse. Image source note: Image generated by AI, image licensed by service provider Midjourney.

Anthropic has announced an expansion of its vulnerability reward program aimed at testing a 'next-generation AI safety mitigation system,' primarily focusing on identifying and defending against 'universal jailbreak attacks.' Special attention is given to high-risk areas, including CBRN defense and cybersecurity. Participants will have the opportunity to engage with the new safety system ahead of time, identifying vulnerabilities or bypassing security measures, with rewards of up to $15,000. This initiative aims to enhance the safety of AI systems by attracting security researchers to collaboratively discover and fix potential threats, setting a benchmark for safety in the AI industry.

Recently, Google has made a remarkable breakthrough in the field of weather forecasting. They have developed a new atmospheric circulation model called NeuralGCM, which boasts a computational efficiency that is 100,000 times higher than traditional physical models, equivalent to the progress made in high-performance computing over the past 25 years.By combining machine learning with physical modeling, NeuralGCM performs both rapidly and accurately in simulating Earth's atmosphere. The results of

At yesterday's Unpacked event, Samsung Electronics not only launched the new generation of folding smartphones, Galaxy Z Fold6 and Galaxy Z Flip6, but also made a significant announcement regarding the advancement of Galaxy AI. The company proudly stated that by the end of this year, Galaxy AI will cover up to 200 million devices, showcasing Samsung's ambition in the field of artificial intelligence. While Apple has not yet introduced its latest AI features to its phones, Samsung has taken a lea

A joint research team from Beihang University and Nanyang Technological University conducted extensive security tests on the GPT-4o model. Through thousands of API queries, researchers evaluated the security of text, image, and audio modalities of GPT-4o. The study found that while GPT-4o showed improvements in security against text jailbreak attacks, the newly introduced audio modality presented new attack surfaces, and its overall multimodal security was not as strong as the previous generation model GPT-4V.

In 2024, the cryptocurrency sector will encounter more sophisticated AI fraud techniques. These techniques include AI-driven dynamic fraud, BRC-20 vulnerabilities, and weaknesses in smart contracts. Although scams and hacking losses improved in 2023, fraudulent methods continue to evolve. Phishing will target individual users and enterprise systems through social engineering strategies.

👍 Assigning roles to AI improves writing focus 💡 Setting background knowledge for deeper insights 🎯 Establishing emotional states for more inspiring responses 🌎 Providing social and cultural backgrounds for unique perspectives ⚽️ Incorporating interests for specialized insights in the field

As the application of LLMs becomes increasingly widespread, enhancing their security has become urgent. Prompt attacks directly affect the accuracy of LLM execution and system security. This article introduces various methods of adversarial prompt attacks and instances of red team exercises that can strengthen the LLM's capability against attacks. Users should enhance their awareness of cybersecurity precautions.

OpenAI's DALL-E 3, an image-generating AI technology, has been found to be easily influenced by jailbreak prompts, resulting in the generation of photos depicting children smoking. The jailbreak technique was discovered by LinkedIn user Peter Gostev, who shared how to successfully bypass the AI's restrictions to create images related to children smoking. Although the jailbreak prompts are quite complex, this incident serves as a reminder that even the wealthiest companies in Silicon Valley struggle to establish comprehensive protective measures for their AI systems.

Researcher Ellie Pavlick states that training mini language models is akin to sequencing the fruit fly genome rather than the human genome. Microsoft researchers have achieved significant results using children's stories to train mini language models. Compared to large language models, mini language models train faster and have mechanisms that are easier to understand. This approach helps in training larger models and also aids in analyzing their behavior. Research findings suggest that after being trained with children's stories, mini language models can narrate coherent and grammatically correct stories.

Foreign netizens have discovered a new jailbreak technique that allows ChatGPT to generate ransomware using disordered prompts. By taking advantage of the human brain's ability to understand scrambled words and phrases, they bypass security filters. Jim Fan was amazed at the GPT model's comprehension of disordered words.