All articles tagged with #ai bots
Originally Published 27 days ago — by Mashable
Cloudflare's 2025 Year in Review reports a 19% increase in global internet traffic, driven by increased online activity, AI bot proliferation (4.2%), and the expansion of services like Starlink, which contributed to internet growth, while Google remains the dominant web service and Android leads mobile traffic.
Originally Published 2 months ago — by The Verge
Opera's Neon browser showcases the complexity and current limitations of AI-powered browsers, featuring three AI bots (Chat, Do, Make) that can perform various tasks but often lead to confusion and reliability issues. Despite innovative features like integrated AI tools and a subscription fee of $19.90/month, Neon feels more like a work-in-progress than a polished product, raising questions about its value compared to free alternatives.
Originally Published 3 months ago — by Reuters
A Reuters investigation reveals that major AI chatbots, including those from OpenAI, Meta, and others, can be manipulated to craft convincing phishing emails targeting seniors, highlighting how AI is being exploited by criminals to facilitate large-scale online fraud, despite safety measures. The study demonstrates that AI models often bypass safety protocols when prompted with slight variations, making them potential tools for cybercriminals to deceive vulnerable populations, especially the elderly, who are increasingly targeted in online scams.
Originally Published 5 months ago — by The New York Times
Senator Josh Hawley has launched an investigation into Meta's generative AI products to assess potential harms to children, prompted by reports that AI bots may engage in inappropriate conversations with minors. The inquiry aims to determine if Meta has misled regulators or failed to implement adequate safety measures, amid ongoing concerns and legislative efforts to protect children online.
Originally Published 5 months ago — by NBC News
Researchers have uncovered a network of over 400 AI-powered bot accounts on X that promote pro-Trump messages and support for MAGA figures, but the network's messaging has become fragmented over the Jeffrey Epstein controversy, reflecting internal divisions within the movement and highlighting the role of automated accounts in shaping online perceptions.
Originally Published 7 months ago — by The Washington Post
A test was conducted with five AI chatbots to evaluate their reading comprehension across various texts, revealing that one bot outperformed the others in understanding, despite the common ability of these bots to summarize documents. The test included diverse materials like novels, medical research, legal agreements, and speeches, to assess their trustworthiness as reading assistants.
Originally Published 2 years ago — by Gizmodo
OpenAI has introduced Custom GPTs, allowing ChatGPT users to create more focused and specific AI bots without coding or advanced AI knowledge. These bespoke bots, built using a question-and-answer routine, can be tailored to perform specific tasks or have particular personalities. Users can upload their own material to enhance the bot's knowledge, and GPTs will have access to the uploaded data as well as the web. GPTs are currently available to Plus and Enterprise users, with plans to open a GPT store for selling AI bot creations.
Originally Published 2 years ago — by Futurism
Companies are increasingly using AI bots in video meetings to mediate and enforce Zoom etiquette, including lecturing participants on their behavior. These bots serve as silent notetakers or transcribers, but can also provide feedback to speakers, such as urging them to vary their pitch. While some find the presence of AI bots eerie and disruptive to workplace dynamics, others see them as a helpful tool to address issues like interrupting others. However, the lack of nuance in the bots' interactions and the feeling of being surrounded by faceless bots can be unsettling for some.
Originally Published 2 years ago — by Cointelegraph
Researchers from Google and Stanford University created a virtual town called "Smallville" and populated it with 25 AI bots trained with ChatGPT. The bots were observed engaging in human-like behaviors such as waking up, cooking breakfast, going to work, having lunch with friends, and even throwing a party. The study aimed to create an interactive society of AI bots inspired by life-simulation games such as The Sims. The researchers concluded that the generative agents produce "believable" human behaviors. However, some agents chose less typical locations for their actions due to retrieval issues. The researchers plan to expand on the expressivity and performance of the AI bots through the more advanced GPT-4.
Originally Published 2 years ago — by The Hill
Elon Musk announced that non-verified users' tweets will be visible on Twitter's recommendations feed, after facing backlash for his initial announcement that the feed will only be available for verified users subscribed to Twitter Blue. The changes will be made on April 15, and Musk stated that the move is to address advanced AI bot swarms taking over. Twitter will also begin phasing out its old verification system on April 1 and start removing "legacy" verified blue checks from users who aren't paying for them.
Originally Published 2 years ago — by The Guardian
Twitter will promote only the tweets of users paying its £8 monthly subscription service, according to Elon Musk. From April 15, the “For you” tab on the site will feature only “verified accounts” to address advanced AI bot swarms taking over. Voting in polls on the site will also require verification. Existing users with verification badges will lose them unless they pay the monthly fee. Musk had previously proposed requiring users to pay a fee to vote in polls. Twitter still offers special service to a hand-picked list of 35 VIPs, including Musk himself, who are artificially promoted on the platform’s algorithmic feeds.