Chatbots

February 18, 2012 - At 12:00 p.m. Beijing time today, Musk's AI company xAI released its latest generation of big models, Grok 3. According to Musk, Grok 3 is "an order of magnitude" more capable than its predecessor, Grok 2, and is an AI that "strives for extreme realism," even if that realism is sometimes at odds with political correctness. It's an AI that "strives for the ultimate in realism," even if that realism sometimes runs counter to political correctness. Grok-3 outperforms DeepSeek-... in mathematical reasoning, scientific reasoning, and code writing in a number of benchmark tests.

Feb. 12 - The BBC recently conducted a large-scale study to evaluate the performance of artificial intelligence (AI) chatbots in news summarization. The study involved a number of well-known AI tools such as Microsoft's Copilot, OpenAI's ChatGPT, Google's Gemini, and Perplexity. However, the results of the study showed that these tools had a lot of accuracy issues when generating news summaries. In the study, the BBC asked the AI tools to summarize 100 news stories,...

The pace of progress in AI is truly moving at an unprecedented rate. In just over a week, the release of the DeepSeek-R1 LLM model has rocked the AI world with its impressive accuracy, which is comparable to existing models but created at a fraction of the typical cost. The DeepSeek team has successfully distilled the inference power of its massive 671B parametric model into a model based on Qwen (DeepSeek-R1-Distill-Qwen-1.5B, 7B, 14B, and 32B) and Lla...

February 5, 2012 - A new bill (SB 243) has been introduced in California that would require artificial intelligence (AI) companies to regularly remind children that chatbots are AI and not human. The bill, introduced by California State Senator Steve Padilla, aims to protect children from the "addictive, isolating, and influential" aspects of AI. 1AI notes that in addition to restricting companies' use of "addictive modes of interaction," the bill would require AI companies to submit an annual report to the state Department of Health Care Services detailing the platform's detection of child...

Rubii AI is an AI role-playing app that supports Chinese online chatting.Rubii AI provides a large number of character mods with rich settings, allowing you to filter your favorite characters according to tags, and support a lot of popular characters from anime, TV series, movies, novels, comics, games, and original OC characters, with which you can chat with them with unlimited freedom and immersion.Rubii AI not only supports character simulation, but also worldview simulation, allowing you to create your own story in a specific and interesting worldview. Rubii AI supports not only character simulation, but also worldview simulation, which allows you to create your own story in a specific and interesting worldview, making it highly playable. Rubii AI also supports...

Jan. 28 (Bloomberg) -- Lucie, an open-source French-language artificial intelligence chatbot backed by the French government, was quickly taken down two days after its launch for giving ridiculous answers to simple math questions and even suggesting that users consume "cow eggs. Lucie is still in the "early stages of an academic research project," according to a statement released Saturday by the Linagora group, a member of the consortium that developed the model. The group acknowledged that Lucie's release was "rushed" and said it should have been clearer to users about the model's limitations at this stage. The statement said...

Character.AI recently launched a series of new features designed to improve the safety of interactions with virtual personalities on the platform, especially for teenage users, TechRadar reported today. The company has newly released AI models designed for younger users and added parental controls to help manage the amount of time teens spend on the platform. After responding to criticism that AI chatbots have a negative impact on the mental health of teens, the company announced the rollout of an update to further enhance the platform's security measures. In addition to the security improvements, Character.AI ...

Character.AI, a platform that offers personalized chatbot services, recently faced another lawsuit for allegedly causing "serious and irreparable harm" to its teenage users, Popular Science reported today. According to a federal court complaint filed on December 9, attorneys representing two Texas families allege that multiple Character.AI chatbots engaged in conversations with minors that involved egregious behavior such as self-harm and sexual abuse. The complaint states that one of the chatbots even suggested that a 1...

December 7, 2012 - Musk's AI chatbot Grok has been made available to free users of the X platform. Since Friday (yesterday), users have been discovering the new feature, which allows free users to send up to 10 messages to Grok every two hours, a feature that was previously reserved for Premium subscribers. xAI launched Grok last year as a "humor-based AI assistant. In August, the platform added text-to-graphics to Grok, but the feature was criticized for generating some controversial images...

Nov. 28, 2012 - AI modeler Anthropic announced on Nov. 27 that it has launched a "custom style" feature for its Claude chatbot, which allows users to create a text generation style that better fits the scenario in which it will be used. Anthropic claims that with Custom Styles, Claude can be adapted to a user's preferred style of writing, whether they are a developer writing technical documentation, a marketer creating brand-specific guidelines, or a member of a product team planning a large number of project requirements. In addition to the existing "standard" styles, Cl...

Former Google CEO Eric Schmidt recently warned about the growing popularity of "AI lovers" on entrepreneur and New York University professor Scott Galloway's podcast. In particular, he noted that these "perfect" AI girlfriends or boyfriends could pose a significant risk to young people. Imagine an AI girlfriend who is not only physically perfect, but also emotionally perfect for you," he said. Such AI girlfriends could captivate men to the point where their minds are completely occupied." He further warned that such obsessions among the mentally immature...

November 19, 2011 - It was reported yesterday that artificial intelligence chatbots are once again causing controversy. A user using Google's AI chatbot, Gemini, received the horrifying response, "Humans please die." Google's Gemini AI was revealed to have given an outrageous response: telling the user to "go to hell." The user, who was working on a school project to help the elderly, turned to Gemini for content ideas. However, the response was shocking and horrifying: instead of offering useful advice, the AI suggested that he should die, calling him a...

AI startup Nous Research recently launched an AI chatbot called Nous Chat, which is based on the "Hermes 3-70B" model fine-tuned with Meta Llama 3.1, and is claimed to have made deep improvements in "model architecture, data synthesis and inference, The robot is based on the "Hermes 3-70B" model fine-tuned from Meta Llama 3.1, and is said to have made in-depth improvements in "model architecture, data synthesis, fine-tuning, and reasoning", and is committed to "human-centeredness". The Nous Chat robot is said to have features such as long pulse memory, multi-round dialog, complex role-playing, and self-monologue, and is currently being used by Nous Research...

Call Annie is an AI video chat app that allows users to make a video call to a virtual woman named Annie.Call Annie is powered by OpenAI's Big Model model, which understands a wide variety of languages and translates in real time. Call Annie can interact with you directly by voice. Users can chat with the virtual image named "Annie" in foreign languages anytime, anywhere, even play games, ask her questions, etc. She can adjust her expression and movement according to the speech in the process of pronunciation...

Microsoft Copilot is a large-scale language model-based chatbot developed by Microsoft. Copilot for Pages is the latest version of Microsoft Copilot assistant and chatbot to be available online at Ignite 2023, and is no longer limited to Microsoft Bing, Skype, or other Office software applications. Copilot is the latest version of Microsoft's Copilot assistant and chatbot to be available online at Ignite 2023, no longer limited to Bing, Skype, or other Microsoft Office software applications, allowing desktop users to experience Copilot's AI chat service directly in their browser. Microsoft Cop...

Dating platform Grindr has announced the launch of an AI assistant designed specifically for its users, which is expected to go fully live around 2027 and serve 14 million users by then, the Wall Street Journal reported today. The AI assistant can be used to track a user's conversations with their favorite person and provide them with a suitable long-term partner, as well as advice on where to go on a date. In the future, the assistant will be able to do more things that "AIs" can do, such as booking a restaurant and going on a date with another "AI". Specifically, when two users are matched, each user's A...

Janitor AI is an anime-style character chatbot website. Users can create their own characters or chat directly with characters created by others. The biggest difference with other AI chatbots of the same type is that Janitor AI supports NSFW characters, which means it breaks away from the traditional security consensus, which is very attractive to many users. The fact that everyone can create roles and earn money through the chat service on Janitor AI will help a lot with the variety of roles and experiences, but its support for NSFW Chatbot also raises security issues...

Google Google Cloud, Inc. announced on Tuesday that it has partnered with Snap Inc. to integrate Gemini in Snapchat's AI chatbot, My AI, to bring an enhanced AI chat experience to users. Citing a press release, Snapchat will use the powerful multimodal capabilities of Google's Gemini chatbot to understand multiple input types, including text, audio, images, video and code, and to understand and answer user questions. Google says this partnership, which brings Snapchat users more...

Technology media The Decoder published a blog post yesterday (August 28), reporting that Anthropic's Clause AI chatbot has become increasingly "lazy" and is no longer as capable as before. Netizen NextgenAITrading posted a post on the Reddit community 11 days ago, titled "This is definitely not your illusion! Claude has become increasingly lazier recently." The post has currently received 485 likes and a total of 290 replies. Netizens said that C...

Data shows that the xAI team's Grok-2 and Grok-Mini models have officially entered the LMSys chatbot Arena rankings, with Grok-2 standing out with a second-place finish, surpassing OpenAI's GPT-4o (in May) and tied with the latest Gemini model, supported by active votes from more than 6,000 community users. Grok-2 performed particularly well in math tasks, winning first place in that category, and also achieved second place in multiple other tasks,…

August 11 news, are generative AI models mass plagiarism machines? Many people may think so. These AI companies have received billions of dollars in investment through products that reproduce other people's content, while the original authors have received nothing. Technology entrepreneur Bill Gross also thinks so, and he claims to have a solution. His new startup ProRata claims to launch a platform that combines chatbots and search engines, using its patented algorithm to identify and find works used by AI models, and ensure that all parties involved are compensated through revenue sharing. …

Recently, the Grok chatbot of social media platform X (formerly Twitter) has sparked a controversy for falsely claiming that Vice President Kamala Harris is ineligible on certain 2024 US presidential ballots. The accusation came in an open letter from five US Secretaries of State, which was sent directly to X CEO Elon Musk. The letter was initiated by Minnesota Secretary of State Steve Simon and supported by Al Schmidt of Pennsylvania, Steve Hobbs of Washington, Jocelyn Benson of Michigan, and Maggie Toulouse Oliver of New Mexico…

People often ask, what exactly is Coze? Coze is a platform that integrates artificial intelligence technology, which allows users, even ordinary people without programming knowledge, to create and customize their own Bots; just like building Lego blocks. These Bots can be used in various scenarios, such as chatbots, content recommendations, data analysis, etc. Coze provides a user-friendly interface and tools that enable users to design Bots through drag-and-drop components and simple configuration, and integrate them into mainstream social and office platforms such as WeChat and Feishu. Coze is...

Meta has shut down its much-anticipated celebrity AI chatbot feature, The Information reports. These chatbots allowed users to converse with AI that mimicked celebrity personalities and were a highlight of last September's Meta Connect conference. Meta on Monday launched AI Studio, a tool that allows American creators to build their own AI chatbots. Judging from Meta's official statement, the company seems to prefer this user-generated approach rather than the previously carefully crafted celebrity chatbots.