 |
|
|
|||||
New York City Mayor Zohran Mamdani faced his first snowstorm as mayor over the weekend wearing a trio of jackets that had his new job title embroidered on the chest and sleeve. One was custom with a message written on the inside collar and typography on the front pulled from New York’s past. Contrary to what you might assume, being elected mayor of New York doesn’t automatically get you access to a wardrobe of customized city agency jackets with “Mayor” embroidered on the outside hanging in the closet for you at Gracie Mansion. Those have to be given or made. [Photo: Adam Gray/Bloomberg/Getty Images] Two of the jackets he wore were given to him: a green fleece from the New York City Department of Sanitation (DSNY), and a black windbreaker from the New York City Emergency Management Department (NYCEM). A third, black, custom Carhartt jacket was personalized at the Brooklyn embroidery shop Arena Embroidery. [Photo: Michael Appleton/Mayoral Photography Office] The custom jacket features “The City of New York” written out in long-limbed serifs originally found on old municipal stationery letterhead from the 1980s and ’90s. The wordmark appears in white on the front right chest. Written inside of the collar, hidden from view of the cameras, is the phrase “No Problem Too Big, No Task Too Small.” View this post on Instagram The typographic style of the “The City of New York” mark is vintage, but it’s also back in vogue. Noah Neary, a senior adviser to Mamdani’s wife, Rama Duwaji, designed the mark, and the style can be seen on items like “New York or Nowhere” brand totes, or even on an “Eric Adams Raised My Rent” shirt from Mamdani’s mayoral campaign. For elected officials, these officially embroidered jackets have become the unofficial uniform at public events when Mother Nature strikes. Surveying fire damage last year in California, for example, President Donald Trump wore a windbreaker with the presidential seal on the front and California Gov. Gavin Newsom wore a quarter-zip with a bear, referencing the state flag. For Mamdani, his jackets signaled common cause with the city’s workers during a deadly storm. Political natural disaster wardrobe choices can easily veer into cosplay, like Republican lawmakers who dress like they’re going to a war zone when they’re just going to Texas. And simply wearing the right clothes to an event is not foolproof. What people remember about Trump’s visit to Puerto Rico after Hurricane Maria in 2017 wasn’t his jacket, but the image of him tossing paper towels and the delay of billions of dollars worth of aid. [Photo: Kara McCurdy/Mayoral Photography Office] Dressing more casually, though, does serve as an important form of visual communication when storms, fires, earthquakes, or other threats arise. You don’t show up to a disaster zone in a suit and tie. For Mamdani, his jackets showed solidarity with a city, its workers, and its citizens during his first snowstorm in office with a custom nod to city history.
Category:
E-Commerce
Have you ever had the experience of rereading a sentence multiple times only to realize you still dont understand it? As taught to scores of incoming college freshmen, when you realize youre spinning your wheels, its time to change your approach. This process, becoming aware of something not working and then changing what youre doing, is the essence of metacognition, or thinking about thinking. Its your brain monitoring its own thinking, recognizing a problem, and controlling or adjusting your approach. In fact, metacognition is fundamental to human intelligence and, until recently, has been understudied in artificial intelligence systems. My colleagues Charles Courchaine, Hefei Qiu, Joshua Iacoboni, and I are working to change that. Weve developed a mathematical framework designed to allow generative AI systems, specifically large language models like ChatGPT or Claude, to monitor and regulate their own internal cognitive processes. In some sense, you can think of it as giving generative AI an inner monologue, a way to assess its own confidence, detect confusion, and decide when to think harder about a problem. Why machines need self-awareness Todays generative AI systems are remarkably capable but fundamentally unaware. They generate responses without genuinely knowing how confident or confused their response might be, whether it contains conflicting information, or whether a problem deserves extra attention. This limitation becomes critical when generative AIs inability to recognize its own uncertainty can have serious consequences, particularly in high-stakes applications such as medical diagnosis, financial advice, and autonomous vehicle decision-making. For example, consider a medical generative AI system analyzing symptoms. It might confidently suggest a diagnosis without any mechanism to recognize situations where it might be more appropriate to pause and reflect, like These symptoms contradict each other or This is unusual, I should think more carefully. Developing such a capacity would require metacognition, which involves both the ability to monitor ones own reasoning through self-awareness and to control the response through self-regulation. Inspired by neurobiology, our framework aims to give generative AI a semblance of these capabilities by using what we call a metacognitive state vector, which is essentially a quantified measure of the generative AIs internal cognitive state across five dimensions. 5 dimensions of machine self-awareness One way to think about these five dimensions is to imagine giving a generative AI system five different sensors for its own thinking. Emotional awareness, to help it track emotionally charged content, which might be important for preventing harmful outputs. Correctness evaluation, which measures how confident the large language model is about the validity of its response. Experience matching, where it checks whether the situation resembles something it has previously encountered. Conflict detection, so it can identify contradictory information requiring resolution. Problem importance, to help it assess stakes and urgency to prioritize resources. We quantify each of these concepts within an overall mathematical framework to create the metacognitive state vector and use it to control ensembles of large language models. In essence, the metacognitive state vector converts a large language models qualitative self-assessments into quantitative signals that it can use to control its responses. For example, when a large language models confidence in a response drops below a certain threshold, or the conflicts in the response exceed some acceptable levels, it might shift from fast, intuitive processing to slow, deliberative reasoning. This is analogous to what psychologists call System 1 and System 2 thinking in humans Conducting an orchestra Imagine a large language model ensemble as an orchestra where each musician an individual large language model comes in at certain times based on the cues received from the conductor. The metacognitive state vector acts as the conductors awareness, constantly monitoring whether the orchestra is in harmony, whether someone is out of tune, or whether a particularly difficult passage requires extra attention. When performing a familiar, well-rehearsed piece, like a simple folk melody, the orchestra easily plays in quick, efficient unison with minimal coordination needed. This is the System 1 mode. Each musician knows their part, the harmonies are straightforward, and the ensemble operates almost automatically. But when the orchestra encounters a complex jazz composition with conflicting time signatures, dissonant harmonies, or sections requiring improvisation, the musicians need greater coordination. The conductor directs the musicians to shift roles: Some become section leaders, others provide rhythmic anchoring, and soloists emerge for specific passages. This is the kind of system were hoping to create in a computational context by implementing our framework, orchestrating ensembles of large language models. The metacognitive state vector informs a control system that acts as the conductor, telling it to switch modes to System 2. It can then tell each large language model to assume different rolesfor example, critic or expertand coordinate their complex interactions based on the metacognitive assessment of the situation. Impact and transparency The implications extend far beyond making generative AI slightly smarter. In health care, a metacognitive generative AI system could recognize when symptoms dont match typical patterns and escalate the problem to human experts rather than risking misdiagnosis. In education, it could adapt teaching strategies when it detects student confusion. In content moderation, it could identify nuanced situations requiring human judgment rather than applying rigid rules. Perhaps most importantly, our framework makes generative AI decision-making more transparent.Instead of a black box that simply produces answers, we get systems that can explain their confidence levels, identify their uncertainties, and show why they chose particular reasoning strategies. This interpretability and explainability is crucial for building trust in AI systems, especially in regulated industries or safety-critical applications. The road ahead Our framework does not give machines consciousness or true self-awareness in the human sense. Instead, our hope is to provide a computational architecture for allocating resources and improving responses that also serves as a first step toward more sophisticated approaches for full artificial metacognition. The next phase in our work involves validating the framework with extensive testing, measuring how metacognitive monitoring improves performance across diverse tasks, and extending the framework to start reasoning about reasoning, or metareasoning. Were particularly interested in scenarios where recognizing uncertainty is crucial, such as in medical diagnoses, legal reasoning, and generating scientific hypotheses. Our ultimate vision is generative AI systems that dont just process information but understand their cognitive limitations and strengths. This means systems that know when to be confident and when to be cautious, when to think fast and when to slow down, and when theyre qualified to answer and when they should defer to others. Ricky J. Sethi is a professor of computer science at Fitchburg State University and Worcester Polytechnic Institute. This article is republished from The Conversation under a Creative Commons license. Read the original article.
Category:
E-Commerce
In December 2025, the Department of Transportation (DOT) put out a call for design concepts for new terminals and concourses at Washington Dulles International Airport. The DOT claimed Dulles had fallen into disrepair and was “no longer an airport suitable and grand enough for the capital of the United States of America.” The agency said it was looking for proposals to either replace the airport’s existing main terminal and satellite concourses or build upon them. It also noted Trump’s executive order calling for classical architecture in federal building projects. Mobile lounges on the tarmac at Dulles International Airport [Photo: carterdayne/Getty Images] A number of firms submitted proposals, including Ferrovial, Phoenix Infrastructure Group, and Alvarez & Marshal Infrastructure and Capital Projects. The submission from Bermello Ajamil & Partners and Zaha Hadid Architects included architectural renderings with a prominent feature that appears to be custom designed for a president who is fond of putting his name on things. [Rendering: Ajamil & Partners/Zaha Hadid Architects, via USDOT] The firms’ proposed terminal design would boast a “grand arch” made of a transparent facade and lettering that reads “Donald J. Trump Terminal.” In some renderings, the name is written out in Trajan, a serif font used by the Trump Organization. In one Reddit thread, commenters criticized the move as “shameless” and brought up Zaha Hadid’s work for authoritarian regimes. [Rendering: Ajamil & Partners/Zaha Hadid Architects, via USDOT] Renderings show the Trump terminal superimposed over the airport’s iconic existing terminal, completed in 1962 with a swooping concave roof and large window sides designed by architect Eero Saarinen. A departures hall in the proposed new building builds on Saarinen’s use of openness and natural light with a continuous skylight over a long-span roof. [Rendering: Ajamil & Partners/Zaha Hadid Architects, via USDOT] Bermello Ajamil & Partners has designed terminals for airports in Miami and Fort Lauderdale. Past projects by Zaha Hadid Architects include Western Sydney International Airport in Australia, Bishoftu International Airport in Ethiopia, and Beijing Daxing International Airport in China. Zaha Hadid Architects did not respond to a request for comment.
Category:
E-Commerce
All news |
||||||||||||||||||
|
||||||||||||||||||