Daily AI Newsletter: Powerful New Chatbot Disappears as Mysteriously as It Arrived

In today’s email:

🎶 AI is Doomed to Fail the Musical Turing Test
🍎 Apple targets Google staff to build artificial intelligence team
✝️ Why Doesn't AI Want to Show Me Jesus Washing the Feet of His Disciples?
🧰 10 new AI-powered tools and resources. Make sure to check the online version for the full list of tools.

Powerful New Chatbot Disappears as Mysteriously as It Arrived

A new AI chatbot known as "gpt2-chatbot" recently made waves on the benchmarking site LMSYS Org, boasting capabilities similar to OpenAI's GPT-4. The origins of this chatbot remain unknown, sparking widespread speculation in the AI community. Initially available for public interaction, the chatbot was quickly pulled offline due to overwhelming traffic, but LMSYS Org has hinted at a future, broader release. Notably, the chatbot demonstrated impressive feats, such as coding a clone of Flappy Bird and solving complex math problems, leading to further intrigue about its capabilities.

The intrigue escalated as comparisons were drawn between gpt2-chatbot and other prominent AI models like Anthropic’s Claude Opus, with Gizmodo’s tests showing similar levels of sophistication. The chatbot managed tasks that even other advanced models struggled with, including solving a math riddle involving inexplicit rules posed by a computer engineering professor. This unique ability has fueled the theory that a major player in the AI industry could be behind its development.

As the AI community continues to buzz about gpt2-chatbot, its mysterious origins and potential affiliations with major AI developers like OpenAI, Google, or Anthropic remain hot topics of discussion. The upcoming weeks are likely to reveal more about the chatbot’s creators and their intentions, potentially signaling the emergence of a new influential force in the AI landscape or the introduction of an advanced new model.

An entirely new way to present ideas

Gamma’s AI creates beautiful presentations, websites, and more. No design or coding skills required. Try it free today.

Amazon just announced Q, the company’s enterprise AI assistant.

Amazon Q is an advanced generative AI-powered assistant designed to accelerate software development and enhance the use of internal data within companies. It offers capabilities such as generating code, testing, debugging, and multi-step planning and reasoning, allowing it to transform and implement code based on developer requests. Additionally, Amazon Q facilitates employee access to a variety of business data, including company policies, product information, and business results, by connecting to enterprise data repositories. This connection enables the summarization of data, trend analysis, and engaging dialogues about the data, streamlining company operations and decision-making processes.

Amazon Q Business, a specific version of the tool, is tailored for enterprise use, providing employees with the ability to generate content, answer questions, and complete tasks securely based on enterprise system data. This tool enhances creativity, efficiency, and productivity by uniting data sources and repositories. Amazon Q also ensures security and privacy by respecting existing governance identities, roles, and permissions, ensuring that if a user is restricted from accessing certain data without Amazon Q, the same restrictions apply when using the tool.

Prominent companies have reported significant improvements after implementing Amazon Q. Smartsheet, for instance, utilized Amazon Q Business to streamline knowledge management and boost productivity, consolidating organizational knowledge into a single AI engine for immediate employee access. Novacomp used Amazon Q to modernize a large coding project, reducing the completion time from weeks to minutes and significantly decreasing their technical debt. Similarly, BMW and the National Australia Bank have leveraged Amazon Q to enhance their operational efficiencies and decision-making processes, further demonstrating the tool’s impact across various industries.

8 Daily Newspapers Sue OpenAI and Microsoft Over A.I.

Eight daily newspapers, including The New York Daily News and The Chicago Tribune, owned by Alden Global Capital, have filed a lawsuit against OpenAI and Microsoft. The lawsuit, filed in the Southern District of New York, accuses the tech giants of using copyrighted articles from these newspapers to train their A.I. systems, such as ChatGPT and Microsoft Copilot, without permission. This has allegedly led to the misuse of the content, including displaying articles behind paywalls and not linking back to the source, thus harming the newspapers' subscription revenue and licensing opportunities.

The lawsuit highlights specific instances where the A.I. chatbots produced by OpenAI and Microsoft have misrepresented or inaccurately cited newspaper content, damaging the publications' credibility. For example, ChatGPT incorrectly stated that The Chicago Tribune recommended a recalled infant product, and claimed The Denver Post published research suggesting smoking could cure asthma, both of which were untrue. The newspapers are seeking a jury trial and compensation for the alleged unauthorized use and damage caused.

This legal action adds to the ongoing debate over the ethical use of publicly available data to train A.I. systems, a practice that has become central to the development of generative A.I. technology. Other publications, like The Financial Times and The Associated Press, have negotiated agreements with tech companies for the use of their content, suggesting a potential path forward amidst these disputes. The case underscores significant concerns about the impact of A.I. on journalism and information integrity.

Other stuff

AI is Doomed to Fail the Musical Turing Test
Yelp is launching a new AI assistant to help you connect with businesses
Why Doesn't AI Want to Show Me Jesus Washing the Feet of His Disciples?
Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
local GLaDOS - real-time interactive agent, running on Llama-3 70B
Samsung’s operating profit soars 930% as AI tailwinds drive demand for memory chips
Sam’s Club’s AI-powered exit tech reaches 20% of stores
AI leaderboards are no longer useful. It's time to switch to Pareto curves.
Malleable software in the age of LLMs
Apple targets Google staff to build artificial intelligence team