At the recent Paris AI Action Summit, OpenAI CEO Sam Altman revealed that OpenAI is willing to engage in deep collaboration with China in the field of artificial intelligence. This statement has garnered widespread attention, particularly regarding his high praise for the Chinese AI startup DeepSeek. He noted that DeepSeek's performance in developing chatbots is 'remarkable', and expressed that the company...

Recently, a study by the BBC has revealed that leading AI assistants often produce misleading and inaccurate content when answering questions related to news and current affairs. The research indicates that over half of the responses generated by four mainstream AI tools—ChatGPT, Gemini, and Perplexity—are deemed to have 'significant issues'. Note: The image is AI-generated, provided by the image licensing service Midjourney, as requested by researchers for these four generative AI tools.

In 2024, global mobile users spent an astonishing $1.27 billion on AI chatbots and art generation tools. This figure represents significant growth compared to the previous two years, with spending on chatbot applications reaching $1.07 billion, reflecting a strong demand and recognition for AI assistants. According to the latest market report, ChatGPT leads in both downloads and consumption in the AI-generated content (AIGC) application space, becoming the preferred choice for users. Following closely is Google's Gemini.

Recently, Nvidia announced the addition of three safety features on its NeMo Guardrails platform, aimed at helping businesses better manage and control AI chatbots. These microservices target common challenges in AI safety and content moderation, offering a range of practical solutions. Among them, the Content Safety service can review content before the AI responds to users, detecting any potential harmful information. This service helps prevent...

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.

Against the backdrop of Trump's return to the White House, top entrepreneurs from Silicon Valley are quietly becoming key technological advisors to the president. Elon Musk, Marc Andreessen, and David Sacks are among those strategizing for Trump, with their most notable point being their strong criticism of AI regulation. This group of tech elites believes that AI chatbots are emerging as a new ideological battleground. Their core concern is that large tech companies may use AI systems to propagate specific political positions, achieving a more covert and effective form of 'regulation'.

According to a report by The Information, OpenAI is planning to launch an in-house browser powered by an AI chatbot to directly challenge Google's search market. This move highlights OpenAI's ambitions in the search domain and is also a significant step in expanding its ecosystem. Sources indicate that OpenAI is in discussions with platforms such as Condé Nast, Redfin, Eventbrite, and Priceline for potential collaborations in travel, food, and housing.

According to a recent report by The Information, Meta is quietly advancing an ambitious plan to develop a dedicated search engine for its AI chatbots. This strategic shift stems from Meta's deep reflection on the significant losses caused by changes to Apple's privacy policies. The search engine project has been under wraps for at least eight months. The company aims to integrate web indexing into Meta AI, thereby reducing its dependence on Google Search and Microsoft Bing.

Recently, a study published in Nature revealed a thought-provoking phenomenon in the development of artificial intelligence: as large language models (LLMs) continue to evolve, they increasingly tend to provide answers with 'full confidence', even when those answers may be incorrect. This phenomenon has sparked broad discussions on AI reliability and the risks of use. The research team, led by José Hernández-Orallo from the Valencia Institute of Artificial Intelligence in Spain, conducted an in-depth analysis of AI models.

Judge Kevin Newsom of the US 11th Circuit Court of Appeals innovatively used chatbots like ChatGPT to interpret the legal term 'ordinary meaning,' particularly concerning the dispute over whether a 'physical restraint' was involved in a gun robbery case. In this case, the judge agreed to uphold the severe sentence for the defendant DeLeon but noted that there may have been past misinterpretations of the definition of 'physical restraint' and emphasized that the term should be understood according to its ordinary, common meaning. Newsom asked GPT-4, Claude, and Gemini about the ordinary meaning of 'physical restraint.'

Replika, an AI chatbot designed to help users build emotional and social connections, has recently come into the spotlight. CEO Eugenia Kuyda stated in an interview with The Verge that some users are even willing to 'marry' their AI companions, believing it brings happiness and reduces loneliness. Kuyda argues that AI companions can be effective tools for emotional support, helping people establish friendships and a sense of connection. Despite facing controversy in the past, such as issues related to users' sexual interactions with their AI partners and incidents of verbal abuse,

Google has conducted a significant upgrade to the AI chatbot Gemini to enhance performance, expand language support, and increase its reach. The Gemini 1.5 Flash version supports 40 languages across 230 countries and regions, with notable improvements in quality and response speed, particularly in reasoning and image understanding. This upgrade includes a 32,000-token context window expansion, support for file uploads, and the display of relevant webpage links below generated answers to improve transparency. Gemini's functionalities will expand into more application areas.

The latest research finds that AI-powered chatbots can pass certified ethical hacker exams, but researchers warn that they should not be solely relied upon for cybersecurity protection. The study was conducted by researchers Prasad Calyam from the University of Missouri and collaborators from Amrita University in India, and tested the leading generative AI tools ChatGPT by OpenAI and Bard (now known as Gemini) by Google.Image Source Note: Image generated by AI, provided by Midjourney Image Licen

Instagram has launched 'AI Studio', allowing creators to create their own versions of AI chatbots. ⭐ The AI created by the creators will primarily appear in Instagram's messaging and will be clearly labeled as AI. ⭐ Meta stated that they aim to enable creators and small businesses to create AI to engage with their communities and customers.

A research team from University College London and Eberhard Karls University Tübingen developed Limbic Access to improve the accessibility of mental health services. Observational studies show that the use of Limbic Access has led to more people utilizing NHS talking therapy services, particularly among minority groups. The system reduces the social stigma surrounding mental health issues through personalized design and guided self-referral processes. During its use across 14 NHS talking therapy service centers in England, self-referral numbers increased by 15%. Non-binary, Asian/British Asian individuals and others...

Various AI chatbots have been launched: OpenAI's ChatGPT, Google's Bard, Microsoft's BingChat, and the recently public-tested Grok by xAI. While people are amazed by the powerful capabilities of generative AI, they also often struggle with its 'hallucinations.' On the second day of Grok's open beta, a security tester, Jax Winterbourne, shared a screenshot on X showing Grok denying a request: 'I'm afraid I cannot fulfill this request because...

Essential AI, founded by former Google executives, has raised $56.5 million in new funding. Essential AI developed an AI technology called 'Enterprise Brain' that performs enterprise functions and automates tasks. The founder of Essential AI authored the paper 'Attention Is All You Need', which outlines the foundation of large language models at Google. Large language models expand AI capabilities across text, images, speech, video, and music.