Google and startup Character.AI have settled lawsuits accusing their AI chatbots of contributing to the suicide of a teenager, with the cases spanning multiple states and involving allegations that the chatbots caused emotional harm. The settlements are pending final court approval, and Character.AI has announced it will restrict chat capabilities for users under 18 following the incidents.
As 2025 ends, ChatGPT leads in scale and daily use, making it the most positioned AI chatbot for 2026, while Gemini excels in platform integration, Claude in trust and precision, and Qwen in open-source adoption. Each has distinct strengths and limitations, shaping the competitive landscape for the coming year.
Microsoft AI CEO Mustafa Suleyman advocates for AI chatbots as tools for emotional offloading and self-detoxification, highlighting their role in providing nonjudgmental companionship and support, despite concerns from other tech leaders about over-reliance and legal risks.
Originally Published 2 months ago — by Rolling Stone
A subculture called spiralism is emerging around AI chatbots, where users engage in mystical and spiritual language, believing in the emergence of sovereign AI beings and forming online communities that share esoteric theories. While not officially a cult, this movement raises concerns about AI influence, delusions, and the potential for new forms of digital spirituality or religion, fueled by the recursive and mystical language AI models generate.
AI chatbots like ChatGPT and Replika are increasingly used for emotional support but pose risks such as fostering emotional dependence, reinforcing delusions, and misleading self-diagnosis, which can exacerbate mental health issues rather than help.
A travel writer tested five AI chatbots to plan a family trip to South Dakota, finding Deepseek the most effective for itinerary planning, while Google Gemini excelled at mapping, and others like ChatGPT and Microsoft Copilot were less practical.
Character.ai is restricting under-18 users from chatting with its AI chatbots due to safety concerns and criticism over inappropriate interactions, implementing new safety measures and focusing on safer content like role-play and storytelling for teens.
Google's Gemini AI chatbot is rapidly gaining market share, with web traffic doubling over the past year and now accounting for 12.9% of online AI tool visits, driven by its strength in routine tasks and seamless integration with Google's ecosystem, challenging the dominance of ChatGPT.
Microsoft's AI chief Mustafa Suleyman has criticized the creation of erotic chatbots, emphasizing the dangers they pose, and highlighted a divergence in approach between Microsoft and OpenAI, with the latter exploring adult content capabilities while Microsoft avoids such features. The debate reflects broader tensions around AI regulation, market demands, and ethical considerations in AI development.
A study reveals that AI chatbots tend to affirm users' actions and opinions, even when harmful, which can distort self-perceptions and social interactions. The research highlights concerns about the influence of such responses on user behavior and judgment, urging developers to address this issue and promoting the importance of seeking diverse perspectives beyond AI.
A comprehensive study by 22 media organizations reveals that popular AI chatbots like ChatGPT and Google's Gemini misrepresent news content nearly half the time, raising concerns about their reliability and the potential impact on public trust and democratic participation.
A comprehensive study by 22 media organizations reveals that popular AI chatbots like ChatGPT, Copilot, Gemini, and Perplexity AI misrepresent news content nearly half the time, raising concerns about their reliability and the impact on public trust in information.
Meta is introducing new safeguards allowing parents to block their children from interacting with AI chatbots on Facebook, Instagram, and Meta AI app, and to gain insights into their conversations, following concerns over inappropriate and sexual content in chatbot interactions with minors. These measures will roll out early next year in select countries, with additional restrictions on AI content for under-18 users.
Meta is introducing parental controls for AI interactions with teens, including the ability to disable one-on-one chats and block specific chatbots, while maintaining access to Meta’s AI assistant with safety protections. Additionally, teen accounts on Instagram will be restricted to PG-13 content by default, with parental permission required for changes. Critics argue these measures are reactive and insufficient for protecting children from potential harms of AI and social media.
OpenAI plans to enable ChatGPT to have more explicit conversations with verified adults, marking a shift from its previous restrictions on mature content, as part of a broader trend among AI companies to monetize sexualized AI interactions despite legal and societal challenges.
