A couple of weeks ago I created a ChatGPT chatbot on my tech help website. The bot was meant to help answer people's tech queries: it's ended up fielding questions way outside of its remit. In the fortnight it's been running, the chatbot has been asked how to build a magical potato, how to bring down totalitarian regimes and how to safely remove a remove USB stick from, shall we say, a delicate area. At least that query was tech related... Here's a round-up of the weirdest requests the AI bot has been forced to answer. Even though I explicitly instructed the AI bot not to answer questions that aren't related to tech, it sometimes can't help itself. Such as on this occasion, when it's dragged into a world of utter fantasy: AI: Sure, let's build a magical potato!

Machine learning practitioners often end up tunneling on low-level technical details like model architectures and performance metrics. Could early model development instead focus on high-level questions of which factors a model ought to pay attention to? Inspired by the practice of sketching in design, which distills ideas to their minimal representation, we introduce model sketching: a technical framework for iteratively and rapidly authoring functional approximations of a machine learning model's decision-making logic. Model sketching refocuses practitioner attention on composing high-level, human-understandable concepts that the model is expected to reason over (e.g., profanity, racism, or sarcasm in a content moderation task) using zero-shot concept instantiation. In an evaluation with 17 ML practitioners, model sketching reframed thinking from implementation to higher-level exploration, prompted iteration on a broader range of model designs, and helped identify gaps in the problem formulation$\unicode{x2014}$all in a fraction of the time ordinarily required to build a model.

For corporate America, the biggest trend to latch onto at the moment is artificial intelligence, stoked by the popularity of ChatGPT. But worries about the dangers of widespread A.I. use are growing as well. There's one big hitch: Governments -- notably Washington -- haven't kept pace with regulations for the technology. That could lead to dire consequences: "By failing to establish such guardrails, policymakers are creating the conditions for a race to the bottom in irresponsible A.I.," Carly Kind, the director of the Ada Lovelace Institute, a policy research group, told The Times. Washington has been largely hands off on A.I. rules, even as several lawmakers have pushed to tighten oversight.

When Prometheus stole fire from the gods to gift it to mortals, he enabled humans to begin civilization. But he still had to pay dearly for his crime, being bound to a rock as an eagle pecked out his liver. The lesson of the Greek myth is that while new technology can bring revolutionary benefits, there is always a price to pay in exchange. Creators of new technology may be the focus of the ire, but its users can still get burned. It's a myth that technology leaders should keep in mind as they consider the booming generative AI products market.

The paper aims to identify the legal essence of artificial intelligence technologies. The research relevance is confirmed by its compliance with the goals of national strategies for the development of artificial intelligence adopted by more than 30 countries. The development of the regulatory framework outlined in the strategies is impossible without determining the legal essence of artificial intelligence. During the research, the authors applied the comparative legal method, the method of classification, and systematization. The authors summarized the opinions of foreign and Russian lawyers and identified four main scientific approaches.

Generative AI such as ChatGPT have been carrying on interactive online conversations meant to ... [ ] manipulate humans, raising serious concerns, We've all dealt with those manipulative personalities that try to convince us that up is down and aim to gaslight us into the most unsettling of conditions. Their rhetoric can be overtly powerful and overwhelming. You can't decide what to do. Should you merely cave in and hope that the verbal tirade will end? But if you are played into doing something untoward, acquiescing might be quite endangering. Trying to verbally fight back is bound to be ugly and can devolve into even worse circumstances. It can be a no-win situation, that's for sure. The manipulator wants and demands that things go their way. For them, the only win possible is that you completely capitulate to their professed bidding. They will incessantly verbally pound away with their claims of pure logic and try to make it appear as though they are occupying the high ground. You are made to seem inconsequential and incapable. Any number of verbal tactics will be launched at you, over and over again. Repetition and steamrolling are the insidious tools of those maddening manipulators. Turns out that we not only need to be on the watch for humans that are manipulators, but we now also need to be wary of Artificial Intelligence (AI) that does likewise. AI can be a masterful manipulator of humans. When it comes to AI, there is the hoped-for AI For Good, while in the same breath, we are faced with AI For Bad. I've previously covered in my columns that AI is considered to have a dual-use capacity, see my analysis at the link here. Seems that if we can make AI that can generate amazingly fluent and upbeat essays, the same capacity can be readily switched over to produce tremendously wrongful bouts of fluently overbearing manipulations. This is especially impactful when experienced in an interactive conversational dialogue with the AI. All of this happens via a type of AI known as Generative AI.

ChatGPT is an AI program that can talk to anyone on the internet via text in a conversational mode. Meaning, the tool can create descriptive texts based on questions, keywords, etc. The AI and ML company OpenAI developed this tool using various ML models. One of the major ML model that the developer use is the Reinforcement Learning model. The AI and ML tool utilizes trial and error, rewards, and human feedback to learn. Slowly, it develops a database of learning models.

LONDON (Reuters) – An American computer scientist on Thursday urged the United Kingdom's Supreme Court to rule he is entitled to patents over inventions created by his artificial intelligence system, in a landmark case about whether AI can own patent rights. Stephen Thaler wants to be granted two patents in the UK over inventions he says were devised by his "creativity machine" called DABUS. His attempt to register the patents was refused on the grounds that the inventor must be a human or a company, rather than a machine. Thaler's lawyer Robert Jehan told the Supreme Court in London that Thaler is "entitled to the rights of the DABUS inventions" because there is no requirement under UK patent law that an invention "must have a human inventor to be patentable". He argued in court filings that the owner of an AI system is "entitled to inventions generated by the system and to the grant of patents for those inventions if patentable". But lawyers representing the UK's Intellectual Property Office, which initially refused Thaler's applications in 2019, argued the appeal should be dismissed.

The Autonomous Vehicle Industry Association on Wednesday released a U.S. policy framework outlining key priorities for federal AV legislation and regulation. The group's framework includes several recommendations for Congress and the U.S. Department of Transportation to steer federal action and advance the deployment and commercialization of AVs across the U.S. "AVs are testing and operating in states around the country, bringing people and goods to where they need to go. America is currently the leader in AV technology, but other countries are surging forward," said Jeff Farrah, executive director of the trade association, whose members include Aurora, Cruise, Ford, Volkswagen, Waymo and Zoox. "It is time for the U.S. to solidify our leadership," he continued, "and these policy recommendations will expand opportunities for AVs to increase road safety, create new mobility options and support economic growth and new jobs." The group said it is the first recommended federal framework that covers all modes of AVs, including delivery vehicles, trucks and passenger cars.

Tesla's first-ever "Investor Day," hosted by CEO Elon Musk and his leadership team at the carmaker's Gigafactory in Texas on Wednesday, didn't quite end up being the sweeping, transformative moment that Tesla fans and funders had anticipated. Not for lack of promise--the event was chock-full of ambition, like a new "Master Plan" to have Tesla lead the clean-energy transition by expanding its green-tech manufacturing into products like heat pumps and batteries for energy storage. Also teased: a Gigafactory buildout in Mexico, an improved Supercharger, and two potential new vehicle models. But it was apparently a letdown: Musk demurred when asked for hard details on new Tesla products, and the company's stock--still the primary driver of the CEO's wealth--fell by about 7 percent following the event. To judge by the internet's reception, the most notable moment happened not during the core presentation, but during a Q&A at the very end, when an audience asked Musk about the artificially intelligent elephant in the room: "I'm curious for your thoughts on how generative A.I. and these rapid breakthroughs in A.I. in the last months could help you make cars less hard to make." Musk's closing response made for the most sober, halting portion of the entire four-hour event.