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.
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.
As governor of South Dakota, one of the least populated states in the U.S., Kristi Noem still made an outsize name for herself nationally using public service announcement campaigns designed to capture attention. The topics of her PSAs have changed dramatically since then.
Before assuming her current Cabinet post as secretary of Homeland Security, the former state lawmaker and member of the U.S. House of Representatives served as South Dakotas governor from 2019 to 2025. In her first year as governor, the state ran a widely mocked anti-drug campaign called “Meth. We’re On It,” followed by “Freedom Works Here,” a workforce recruitment campaign in which she was featured prominently.
Once elevated to DHS secretary, Noem continued to utilize public funds for commercials promoting her particular brand of political communication, including a 2025 campaign in service of President Trump and his border and immigration policies.
View this post on Instagram
The nearly $1.4 million “Meth. We’re On It.” campaign ran on TV, billboards, and online (via the now defunct website onmeth.com), and it caught plenty of grief for its ambiguous tagline. Noem defended it at the time, writing on social media, “Hey Twitter, the whole point of this ad campaign is to raise awarenes. So I think thats working.”
“Meth. We’re On It.” was made to combat a real problem in the state, as South Dakota ninth graders tried meth at twice the national average, according to the creative brief for the campaign. Ultimately, it saw some success. By 2020, 1,072 people had clicked the find treatment link on the campaign’s website, 184 people called or texted the campaign’s help line, and 44 were referred to treatment. “Meth. We’re On It.” would become a finalist in the public health category for the Shorty Awards, a social media and digital advertising industry awards ceremony.
In spots for Freedom Works Here,” a South Dakota workforce recruitment campaign that aired nationally in 2023 and 2024, Noem dressed as a law enforcement officer, welder, and dentist as a play on the fact that there werent enough people to fill the state’s job openings, so she was doing them herself. As of July 2023, more than 3,500 people had applied to the program, Noem’s office said at the time. Yet the campaign was criticized as self-serving by some Republican state lawmakers.
At DHS, Noem was the face of the biggest political ad of 2025. The agency spent upwards of $50 million of taxpayer funds to air the spot, in which Noem both thanked Trump “for securing our border, deporting criminal illegal immigrants, and putting America first” and called directly for people in the U.S. illegally to leave. Though DHS denied it was a political ad, it sure looked and sounded like one, with B-roll pulled straight from the tropes of Republican attack ads about border security, like shots evoking crime, drugs, trafficking, and chaos at the border.
Through her political career, Noem has appeared in PSAs for vaccines and storm preparedness, though they didn’t receive the same multimillion-dollar budgets as her more controversial ads. She also starred in a video during last year’s government shutdown intended to be shown at airport security checkpoints that blamed Democrats for related closures and service slowdowns (although many airports did not air it).
There is growing bipartisan pressure for Noem to step down or be removed from her DHS post following the fatal shootings of Minneapolis residents Renee Good and Alex Pretti by federal immigration and border protection officers. Republican Senators Lisa Murkowski of Alaska and Thom Tillis of North Carolina have called for her to resign, and House Democrats have threatened to begin impeachment proceedings against her if Trump doesn’t fire her.
Noem built a national profile in part by using her public office as a platform, but from her time in Sioux City to her days in Trumps administration, that platform became less civic and more brazenly partisan.
The anti-drug campaign Noem once defended caused eyerolls and snickers, but at least it was also the catalyst for more than a few calls to a hotline set up to help people facing addiction. Contrast that to 2025s “The Law,” in which she made herself the face of an immigration and border enforcement agenda that’s growing increasingly unpopular with the American public.
Noem’s recent PSA appearances indicate the value the Trump administration places on government as showbiz, and that for Noem, public office is theater.
For designers of the built environment, it’s necessary to take a long view. Years or even decades can go into the design and construction of a single project, and the best built projects can stand for centuries.
But the business of designing buildings is also subject to the upheavals and uncertainty of any given moment, including this very tumultuous one. Looking ahead to the (relatively) short-term future of the next year, Fast Company asked architects from some of the top firms working in the U.S. and around the world to predict the biggest forces shaping the industry this year, and the potential bright spots they might see.
Here’s the question we put to a panel of designers and leaders in architecture:
What challenges do you see architects tasked with solving in 2026, and what are potential new opportunity areas?
Collaboration is key
Affordable housing and supporting community resources are in crisisprojects that deliver proximity to public transportation, social infrastructure, and offer cultural resources such as restaurants and entertainment will be in high demand. Understanding the role a building plays within a broader community is a vital part of the design process that is often lacking. Collaboration needs to extend beyond cities and design teams to integrate community needs.
This year will bring many of the same challenges we have already seen: more pressure to deliver projects faster while maintaining the quality of the work, understanding what a high-performance building actually means, and streamlining public agency approvals. The latter is an area where AI would be a valuable tool to support innovation and efficiency. There is also a growing opportunity for greater partnership and collaboration with academia and architectural practice. It is important that there is heightened collaboration between the two, particularly because the skill sets of architects are expanding [to include] different job descriptions and needs.Nick Leahy, co-CEO and executive director, Perkins Eastman
Resilience is a given
2026 is the year when designing for resilience becomes a given. Innovation will be as much about systems as function, form, and aesthetics. We will think more about embodied carbon, and derive ways to deliver low-carbon buildings without cost premiums. Clients will no longer accept “green is more expensive.”
Opportunities: Reuse and reinventionthe second life of a building or district. Conversion of outmoded office buildings to residential and hotels where practical and possible, particularly with older, charming office stock in places where people want to live. Meanwhile, new office buildings will be A++ “luxury,” designed with new forms of amenities centered on wellness and socialization. In the suburbs, malls can become places where mixed-used districts arise, transformed into incubator or civic spaces, designed around health and wellness.
Parking lots can be filled with characterful streets and special 24/7 precincts. Workforce housing will also be a big opportunity that fills the gap between luxury and market rate, while data and energy projects will be relevant and exciting for architects not for their novelty but rather for the spatial intelligence and thoughtful planning required in their successful realization.Trent Tesch, principal, KPF
Sustainable design is harder than ever
One of the most significant challenges facing the U.S. building market in 2026 will be maintaining momentum for sustainable and regenerative design solutions amid economic and policy headwinds. The U.S. construction market has always been driven by a first-cost first mentality, while sustainable design has held its promise of return on investment in the long life cycle of buildings.
The hurdle has always been there, but now the bar is even higher with changes to the Energy Star program, the cutting of federal grants for clean energy, reductions to climate resilience programs, and more. So, architects and designers must move beyond purely ROI and well-being conversations to demonstrate how sustainability mitigates risk, ensures compliance, and drives long-term financial resilience.David Polzin, executive director of design, CannonDesign
Economic headwinds
At PAU we are continuing to incorporate artificial intelligence in aspects of our workflow, but only to augmentnever to replaceour teams talent and judgment. In 2026, architects will probably continue to face economic headwinds. The strong pace of firm consolidation through mergers and acquisitions continues, leaving the question of whether someday it will largely be a discipline split between boutique practices and behemoth corporations.Vishaan Chakrabarti, founder, PAU
More than just buildings
In 2026 climate volatility, housing inequity, infrastructural breakdown, and economic uncertainty will no longer be background conditions but active forces shaping every decision an architect makes. We will be asked to do more than deliver buildings; we will be expected to repair trust in systemspolitical and economicthat have too often failed communities and the environment.
We must navigate these higher expectations, delivering projects with tangible social, environmental, and economic benefits while grappling with tighter timelines and fewer resources. The central challenge will be remaining responsible to both environmental and civic ideals within delivery models that are not designed to reward either.Claire Weisz, founding principal, WXY architecture + urban design
Better decisions, earlier
Architects are working in a moment where pressure is coming from all sides; climate risks are intensifying, housing affordability remains unresolved, and the industry is still constrained by limited labor and capacity. At the same time, clients increasingly expect early, data-backed answers that show how a design will meet sustainability goals and deliver on long-term building performance outcomes.
The challenge is no longer just designing well but navigating increasing complexity and trade-offs without slowing projects down. This is driving the need to remove fragmentation of information across teams and project phases. The defining challenge that architects and designers will need to solve for in 2026 is making confident, defensible decisions early, when they have the biggest impact on [how] a projects environmental, cost, schedule, and performance outcomes are determined.Amy Bunszel, EVP of architecture, engineering, and construction solutions, Autodesk
They lie. Repeatedly. Shamelessly. They lie even when the truth would be easier. They lie when the lie can easily be debunked. They lie to dominate, confuse, and assert control. They treat contradiction as an attack and disagreement as betrayal. These are defining traits of narcissistic leadership.
Strangely enough, in politics and in organizations alike, we keep rewarding narcissistic leaders by giving them more power. We promote them, fund them, vote for them, excuse them, and normalize their behavior, even when there are unmistakable warning signs that should stop us from doing so.
It is obvious that narcissists seek power. The big (and more burning) question is: Why do we keep giving it to them?
{"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2026\/01\/PhotoLVitaud-169.jpg","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2026\/01\/PhotoLVitaud-11.jpg","eyebrow":"","headline":"\u003Cstrong\u003ESubscribe to Laetitia@Work\u003C\/strong\u003E","dek":"Women power the worlds productivity its time we talked more about it. Explore a woman-centered take on work, from hidden discrimination to cultural myths about aging and care. Dont miss the next issue subscribe to Laetitia@Work.","subhed":"","description":"","ctaText":"Learn More","ctaUrl":"http:\/\/laetitiaatwork.substack.com","theme":{"bg":"#2b2d30","text":"#ffffff","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#3b3f46","buttonHoverBg":"#3b3f46","buttonText":"#ffffff"},"imageDesktopId":91472264,"imageMobileId":91472265,"shareable":false,"slug":""}}
We choose narcissists when were anxious
Narcissism is often confused with confidence, ambition, or charisma. In reality, pathological narcissism is defined by grandiosity, a constant need for admiration, low empathy, intolerance of criticism, and a tendency to instrumentalize others.
At high doses, narcissism is deeply corrosive. Highly narcissistic leaders take greater risks, manipulate more freely, break rules more readily, and do not learn from failure. They externalize blame, rewrite history, and prefer loyal sycophants over competent professionals.
As organizational psychologist Adam Grant has argued, we are rarely naive about narcissistic leaders. Most of the time, we recognize them quickly. They boast. They monopolize attention. They perform outrage. They lie openly and repeatedly. We see itand we still choose them.
One of the main reasons is that chaos makes us crave certainty. In moments of crisiseconomic instability, war, technological disruption, climate anxietywe mistake loud confidence for competence. Nuance feels weak. Complexity feels unbearable. Fear narrows our tolerance for ambiguity. It makes us vulnerable to leaders who promise control, simplicity, and absolute answersno matter how fictional those answers may be.
Seen through this lens, Donald Trump is not really an anomaly. He is a symptom. His constant lying, grandiosity, and contempt for institutions are extreme, but the underlying dynamic is familiar. The same behaviorson a smaller scaleare rewarded every day in companies, startups, media organizations, and public institutions around the world.
7 Things We Must Change If We Want Fewer Narcissistic Leaders
If narcissistic leaders keep rising, it is because our systems keep selecting and protecting them. Changing outcomes requires changing the rules of the game. Here are seven shifts that matter.
1. Stop confusing visibility with value
Narcissistic leaders thrive on attention. They dominate meetings, interrupt others, and flood the space with what appears to be certainty. In too many environments, visibility is mistaken for contribution. To counter this, organizations must actively redesign how influence is expressedby limiting airtime and prioritizing written input, for example. Value should be measured by clarity created, not noise produced. Treating visibility as value creates a moral hazard: Those least constrained by doubt gain disproportionate influence.
2. Make lying costly
Narcissists lie because it works. Lies are tolerated, minimized, or reframed as communication style. This tolerance is fatal. False statements must be corrected publicly and promptly. Repeated dishonesty should carry clear reputational and career consequences. Treating truth as optional corrodes institutions fast. The longer a lie goes unchallenged, the more it signals that reality is negotiableand that power, not truth, sets the terms.
3. Evaluate leaders on collective outcomes
Narcissistic leaders often look impressive on individual metrics while quietly hollowing out their teams. Measuring leadership without accounting for turnover, burnout, disengagement, and loss of trust is profoundly wrong. Collective intelligence, psychological safety, and learning capacity must be treated as core performance indicatorsnot soft, secondary concerns. If results are achieved at the expense of trust, retention, and learning, they represent short-term extraction rather than sustainable performance.
4. Stop rewarding the will to power
Aggressively wanting power is not proof of leadership potential. In fact, narcissistic personalities are statistically more likely to self-nominate, campaign for authority, and pursue promotion relentlessly. Systems that equate ambition with suitability all but guarantee poor outcomes. Leadership selection should deliberately include capable individuals who do not seek power for its own sakeand should treat excessive self-promotion as a risk signal.
5. Institutionalize dissent
Narcissistic leaders fear contradiction and punish it, directly or indirectly. That is why dissent cannot rely on individual bravery alone. Organizations must structurally protect disagreement through formal devils advocate roles, strong whistleblower protections, and explicit rewards for surfacing bad news early. A leader who cannot tolerate dissent is fundamentally dangerous. Disagreement should be seen as a contribution to intelligence.
6. Redefine charisma
Charisma is too often equated with dominance, theatrical confidence, and verbal force. But sustainable leadership can look different: calm authority, restraint, curiosity, and the ability to change ones mind in light of new evidence. As long as we glamorize the worst kind of strong personalities, narcissistic leaders will continue to thrive. Our dominant definition of charisma is also deeply gendered. Traits coded as charismaticassertiveness, verbal dominance, emotional detachment, physical presencemap closely onto traditionally masculine norms, while behaviors more often associated with women (like listening) are systematically undervalued.
7. Address the root cause: Fear
Narcissistic leaders rise fastest in anxious systems. When people feel unsafeeconomically, socially, psychologicallythey outsource certainty to those who project it most loudly. Reducing precarity, increasing fairness, and building real psychological safety are not just mral imperatives. They are structural defenses against narcissistic leadership.
Narcissistic leaders do not seize power alone. They are enabledby our fears, our metrics, our myths about leadership, and our reluctance to confront uncomfortable truths. If we want different leaders, we must become different selectors. The problem is not that narcissists exist. Its that we keep mistaking them for leaders.
{"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2026\/01\/PhotoLVitaud-169.jpg","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2026\/01\/PhotoLVitaud-11.jpg","eyebrow":"","headline":"\u003Cstrong\u003ESubscribe to Laetitia@Work\u003C\/strong\u003E","dek":"Women power the worlds productivity its time we talked more about it. Explore a woman-centered take on work, from hidden discrimination to cultural myths about aging and care. Dont miss the next issue subscribe to Laetitia@Work.","subhed":"","description":"","ctaText":"Learn More","ctaUrl":"http:\/\/laetitiaatwork.substack.com","theme":{"bg":"#2b2d30","text":"#ffffff","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#3b3f46","buttonHoverBg":"#3b3f46","buttonText":"#ffffff"},"imageDesktopId":91472264,"imageMobileId":91472265,"shareable":false,"slug":""}}
Two years ago, the last pay phones were disconnected in Rochester, New York. But a group of volunteers has started bringing a handful of phones backand making them free to use.
Called the GoodPhone Project, the effort is aimed at people who still dont have reliable access to a mobile phone, including those experiencing homelessness. As pay phones have disappeared, alternatives have been hard to come by.
A lot of community centers, and especially the Monroe County libraries, were being inundated with people asking to use their front desk phones, says Eric Kunsman, one of the volunteers behind the project.
[Photo: Eric T. Kunsman/GoodPhone]
Theres a clear need: The handful of phones that the GoodPhone Project has installed each get hundreds of uses per month. Around 20% of the calls go to social services. (Calls have a 20-minute limit unless theyre made to social service organizations, since that often involves a long wait on hold.)
The upcycled phones use voice over internet protocol (VoIP) technology and allow users to set up their own voicemail extensions, so its possible to use the number when they apply for a job.
Kunsman, a photographer who teaches at the Rochester Institute of Technology, spent years photographing pay phones in the area as they were slowly taken away. He soon realized that despite seeming like relics, the phones were still in use. He partnered with colleagues Rebekah Walker, a digital librarian, and researcher Janelle Duda-Banwar to map out the phones locations. The last phones to survive were in the poorest areas.
[Photo: Eric T. Kunsman/GoodPhone]
When the Frontier phone company removed the last phones, the group decided to do something. They found a Los Angeles-based company that still installs pay phones, called Littlejohn Communications, and converted old phones to add VoIP and make it clear that theyre free to use. Six have been installed so far, all in neighborhoods that had the most need. One of the phones is solar-powered, since its in a location that didnt have access to electricity.
The project is relatively inexpensive. An old pay phone costs around $350. (Kunsman tried to acquire Frontiers old phones in Rochester before they were scrapped, but didnt succeed.) The digital device to convert it costs $50, and operations cost around $40 per month. Kunsman has also received some donations of equipment, and says he currently has around 200 old pay phones sitting in front of his photography studio.
[Photo: Eric T. Kunsman/GoodPhone]
He hopes that the city or county can take over the project as a public service and expand it. Im a photographer, Kunsman says. If Im still doing this in five years, we failed in some way.
Previously there were around 1,400 pay phones in Rochester. Roughly 20 years ago, there were 2 million nationally; by 2016, that number had dropped to 100,000. At that point, the Federal Communications Commission stopped tracking the number that were left.
Kunsman wants to make a guide for other communities that want to replicate the process. Since December, he says hes heard from groups in seven other cities that recognized a similar need. “If a photographer, a social sciences librarian, and others can do this, it’s actually a lot easier than it seems,” he says. “You just need to have the time.”
Is drinking coffee good for your health? The answer appears to be yes. Quite a lot of research shows that coffee drinkers stay mentally sharper and may live longer than those who dont. Thats welcome news to many entrepreneurs and business leaders who depend on coffee to stay alert and productive.
More recent research adds a twist to the welcome news that coffee is good for you. To get coffees life-extending benefits, make sure to drink it in the morning. Dont keep on guzzling the stuff all day long.
Thats the finding of a massive study from Tulane University in New Orleans. Researchers examined data from the National Health and Nutrition Examination Survey (NHANES), which had 40,725 U.S. adult participants and ran from 1999 to 2018.
They wanted to know whether the timing of coffee consumption affected its health benefits. So they divided the cohort into three groups. Thirty-six percent were morning-type coffee drinkers. They did most of their coffee drinking between 4 a.m. and noon, and barely drank any coffee after that. Fourteen percent were all-day-type coffee drinkers who drank coffee throughout the day. And 48 percent were defined as non-coffee drinkers.
The researchers did statistical analysis to determine the mortality rates for each of these groups after about 10 years.
Morning coffee drinkers live longer
Because the NHANES study collected very detailed information on its subjects, the Tulane research team was able to adjust for a wide range of other lifestyle and health factors that could affect longevity. These included everything from body-mass index, to healthy or unhealthy eating habits, to cholesterol levels, and even whether people had trouble sleeping. After accounting for all of these other factors, the researchers consistently found that the morning coffee drinkers had lower mortality than those who drank coffee all day, or not at all.
Taking a closer look, they found that morning coffee drinking particularly seemed to lower the risk from cardiovascular disease. Cardiovascular disease was the leading cause of death for study participants, and is the leading cause of death for Americans overall.
Why should time of day matter so much to getting the benefits of coffee consumption? The researchers found two possible explanations. First, consuming coffee in the afternoon or evening may disrupt circadian rhythms, they wrote. An earlier study had found that heavy coffee drinking in the afternoon or evening interferes with melatonin production at nighttime. Some evidence suggests that low levels of melatonin are associated with higher oxidative stress levels, blood pressure levels, and [cardiovascular disease] risk.
Research has repeatedly shown that not getting enough high-quality sleep is bad for both your health and cognitive function, and even for your leadership abilities. So it seems highly plausible that if drinking coffee later in the day affects the quality or quantity of your sleep, those ill effects could outweigh any benefits you get from coffee drinking.
Coffee is anti-inflammatory
The second possible explanation is that coffees benefits mostly come from its anti-inflammatory properties. Again, this would make sense because many of the worst human ailments, including cardiovascular disease and cancer, are associated with inflammation. The researchers note that inflammation within the body seems to have its own circadian rhythm. Its highest in the morning and lowest around 5 p.m. According to this theory, morning coffee drinking brings the most benefits because it attacks inflammation when its at its worst.
Whatever the explanation, the takeaway is clear. If you love coffee, then great! Drink all you want. But if you want to enjoy coffees health benefits, drink it from when you wake up until noon. After that, consider switching to water, sparkling water, or tea.
Minda Zetlin
This article originally appeared on Fast Company‘s sister publication, Inc.
Inc. is the voice of the American entrepreneur. We inspire, inform, and document the most fascinating people in business: the risk-takers, the innovators, and the ultra-driven go-getters that represent the most dynamic force in the American economy.
Last March, President Donald Trump signed an executive order declaring that the government would launch a strategic reserve of Bitcoin. For the crypto industry, the move was a major win, the next step in its quest to normalize digital assets.
Now, nearly a year later, the amount of Bitcoin held by the U.S. government does seem to be growing, but the federal government also seems somewhat reluctant to talk about about if, and how, the stockpile will actually be set up.
As of January, the U.S. government appears to have amassed about $29 billion worth of Bitcoin, many from seizures that follow criminal investigations, according to a new analysis by Chainalysis, a blockchain data firm. Thats up nearly 50% from May of last year, when the group last conducted a study of government-linked crypto wallets.
Those [BTC] numbers continue to go up over time, Eric Jardine, Chainalysiss head of research, told Fast Company. That stockpile is smaller than some private firms also amassing crypto, he explained, but the current total for the U.S. government is quite sizableas big, if not bigger, than every other government.
The growing reserves align with Trumps executive order, which stated there was a strategic advantage to building up the American governments cryptocurrency troves because, like gold, theres a fixed supply. The White House suggests that building up a supply of Bitcoin, like any “resource,” is good for the national interest, though there are forceful criticisms of that notion.
Still, for all the initial fanfare, the Treasury Department has since been relatively quiet about its progress on moving forward with the reserve. While the government does seem to have begun to holdrather than sell offseized Bitcoin, federal agencies mostly ignored Fast Companys requests for comment on how theyre actually enacting the terms of the order. One source in Treasury Department circles said that theres been radio silence when it comes to the stockpile.
In fact, it seems like the reserve may be facing some legal hurdles. As Fast Company reported the story, Patrick Witt, a White House staffer working on crypto issues, indicated on a crypto-friendly podcast that legal conversations about setting up the reserve were still ongoing. That one isit’s interesting, he said. It seems straightforward, but then you get into some obscure legal provisions, and why this agency can’t do it, but actually, this agency could. We’re continuing to push on that. It is certainly still on the priority list right now.
Making the executive order a reality
Executive Order 14233 instructed the Treasury Department to set up offices to manage both a Strategic Bitcoin Reserve and a United States Digital Asset Stockpile, an office to handle blockchain-based assets other than Bitcoin. According to the order, the Treasury Department was supposed to manage both stockpiles and begin looking for ways the government could potentially acquire more Bitcoin without increasing costs for taxpayers. The order also restricted government agencies from selling or getting rid of digital assets, except in a limited set of circumstances.
In June, Tyler Williams, the Treasury Department’s counselor for digital assets, briefly mentioned the stockpileaccording to minute meetings from the Financial Stability Oversight Council, which is housed within the departmentbut provided few details. A policy report from June also discussed the reserve and noted that the Treasury Department had sent considerations to the White House about the reserve and would move forward with the next steps, including looking at ways to actually hold crypto in custody.
Chainalysis looked at crypto addresses that seem to be associated with the government, calculating they held about $29 billion worth of crypto. The company noted there might be some consolidation of Bitcoin accounts, but didnt say which agencies might be currently holding them.
Its not clear if further progress has been made on developing a Treasury-operated stockpile. Earlier this month, Bitcoin Magazine suggested that the U.S. Marshals Service may have even sold government-seized crypto, prompting an outraged post on X from Sen. Cynthia Lummis (R-WY), one of the most pro-Bitcoin legislators in Congress. Witt, from the presidents council of advisers for digital assets, later said on X that hed confirmed that the wallet in play had not been liquidated, as per the executive order. A U.S. Marshals spokesperson told Fast Company: The reporting about the sale of that wallet was in error. They did not fact-check. The Bitcoin is still being held, as per direction of the executive order.”
The reserve no one will talk about
Still, the government seems otherwise reluctant to discuss the reserve. When asked about the state of the stockpile at the World Economic Forum’s annual meeting, held a week ago in Davos, Switzerland, Treasury Secretary Scott Bessent only told reporters: The policy of this government is to add seized Bitcoin to our digital asset reserve after the damages are done. … Our view was first you have to stop sellingwhich we have doneand then we can add the assets and asset forfeitures.
The Treasury Department has not responded to multiple requests for comment from Fast Company regarding more details on the stockpiles operations, and its not clear if the department has actually set up any offices, as the order stipulates.
Witt, meanwhile, has recently hinted that there are still ongoing discussions on how to, legally, make the reserve actually work. During a podcast interview, he mentioned good engagement with a team led by Stephen Miller, White House deputy chief of staff for policy. I think with some of the latest kind of developments and things that we’ve learned, and engagement from general counsels and different agencies, he said, [they have] some good guidance on where we can move out on this executive order of the president, and can do so in a legally sound way. So more to come on that.
Notably, the executive order also established responsibilities for various federal agencies, which are supposed to communicate with the Treasury Department about Bitcoin and other digital assets they might have on hand. Deadlines for those updates have long passed. Stil, only one of more than a dozen federal agencies contacted by Fast Company commented on what they specifically had done to meet the executive orders requirements. That was the Secret Service, which, in addition to protecting the president and foreign diplomats, has a cybercrime team.
The U.S. Secret Service is compliant with the reporting standards set forth in the executive order, said Alexandria Worley, a USSS spokesperson. A majority of the U.S. Secret Services forfeited digital assets belong to victims of criminal activity, and one of the agencys primary investigative goals is to recover and return those assets to their rightful owners. Worley said that the amount of crypto retained by the government is nominal.
Coinbase, which currently has a contract to hold crypto for the U.S. Marshals Service, did not respond to a request for comment about whether its also holding a U.S. stockpile. Neither did Kraken or Gemini, which also offer services for maintaining crypto. Fast Company was not able to identify any solicitations from the Treasury Department that mention the strategic Bitcoin stockpile, and Sen. Lummiss office also ignored multiple requests for comment.
Pro-crypto groups are still rooting for the stockpile, though, and the validation it offers. Hailey Miller, an executive director for the Digital Chambers policy group, said that the reserve can become not just a balance sheet item, but a pillar of U.S. economic and technological competitiveness.” Ji Hun Kim, CEO of the Crypto Council for Innovation, similarly told Fast Company that the digital asset reserve showcases policymakers understanding that digital assets have a crucial place.
Whenever my wife and I go to watch a movie together, lately we tend to pick a new theater close to where we live that’s called 109 Cinemas Premium Shinjuku. There are reclining seats, you get free popcorn in a chill lounge when you arrive, and the supposedly best-in-Japan sound system was tuned by the late music legend Ryuichi Sakamoto. Whats not to like?
But when we went to see 28 Years Later: The Bone Temple this past weekend, we realized it was only showing in the auditorium dedicated to ScreenX, a fairly new format that has been picking up some steam of late. Id heard of it before but I hadnt ever seen it for myself, so I was happy to check it out in the spirit of technological open-mindedness.
{"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/12\/multicore.png","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/12\/multicore-mobile.png","eyebrow":"","headline":"\u003Cstrong\u003E\u003Cstrong\u003ESubscribe to Multicore\u003C\/strong\u003E\u003C\/strong\u003E","dek":"Multicore is about technology hardware and design. It\u0027s written from Tokyo by Sam Byford. To learn more visit \u003Ca href=\u0022https:\/\/www.multicore.blog\/\u0022\u003Emulticore.blog\u003C\/a\u003E","subhed":"","description":"","ctaText":"SIGN UP","ctaUrl":"https:\/\/www.multicore.blog\/","theme":{"bg":"#f5f5f5","text":"#000000","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#000000","buttonHoverBg":"#3b3f46","buttonText":"#ffffff"},"imageDesktopId":91454027,"imageMobileId":91454030,"shareable":false,"slug":""}}
Three screens in one
The X in ScreenX appears to stand for expanded or extended, because ultimately, what you get is three screens in one. Youre mostly just watching a regular 2D movie screen in front of you, but the footage spills onto the sides of the theater for a 270-degree view.
ScreenX was developed by South Korean theater chain CJ CGV, a subsidiary of one of the countrys largest conglomerates. CJ was also behind the 4DX format that competes with other 4D systems like D-Box and MX4D, with various theater chains adopting one or the other.
While ScreenX has been niche in most of the Western world to date, leading American chain AMC struck a deal with CJ last year to add 25 ScreenX auditoriums and 40 4DX screens across the U.S and Europe. Domestic competitor Cinemark also increased its ScreenX footprint, with plans to add 18 auditoriums in the U.S. over 2025 and 2026.
Thats to say that you might well come across it soon without knowing what youre in for, just as I did. And I really wasnt sure what to expect when going into this showing of The Bone Temple.
The combination of Nia DaCosta as director and Sean Bobbitt as cinematographer didnt make me think this would be a movie that was sloppy about its frame composition. And yet clearly, it couldnt actually have been filmed at such an ultra-wide aspect ratio. Where was the footage on the sides even coming from?
A little distracting
The first thing I noticed about the ScreenX footage is that its actually quite easy to ignore. We had good, centrally located seats, but the screens on the left and right walls were significantly dimmer than the central screen. Thats probably for the best, because it allows you to focus on the actual movie while the extended area serves as added ambience.
On the other hand, Id estimate that there was only about an hour of actual ScreenX footage, or roughly half of the movie. Generally, it tended to be used for outdoor scenes, while the side panels were switched off for interior scenes or tighter shots.
This back-and-forth could be a little distracting; the side panels faded in and out gradually, but the lights on the projectors themselves flicked on and off with every transition. Overall, though, I thought the ScreenX scenes were quite effective in The Bone Temple. At times when characters were running away from infected humans, for example, you got a better sense of how they were being stalked through the forest, with the shaky camera movements placed into context by the wider perspective.
CJ says that it works with directors to help create ScreenX footage in post-production, making use of unused second-unit shots, alternate angles, and CGI extensions. Its hard to tell exactly what youre looking at when you concentrate on the extended footage, not least because it tends to be quite blurryas youd expect from any lens pushed to that extreme. But I didnt see anything particularly jarring or low-quality. Even though I was looking at each screen from a different angle, the transitions were seamless.
A picky viewer
I am quite picky about the theaters I watch movies infor example, last year I made a point of going to Tokyos only true 1.43:1 IMAX theater to see Sinners and One Battle After Another. I would not describe ScreenX as a transformative moviegoing experience on that level. But I think it could have its place.
For movies like The Bone Temple, which is unlikely to be filmed with formats like IMAX in mind but could still benefit from a more immersive presentation due to its intense action, I could see ScreenX being a solid option. Unlike 3D or 4D, there isnt really much compromise or distractionyou can just watch the main screen as usual and absorb the extra footage without actually paying attention.
At the same time, its hard to imagine a movie in which ScreenX would ever be the definitive way to watch it. No director is truly framing their movie with ScreenX in mind; whats in the regular composition is always going to be the actual movie. No one will be watching the ScreenX footage once it leaves theaters.
As such, Id describe ScreenX as a gimmick, but not a particularly destructive one. I dont feel like it had a negative impact on my appreciation of The Bone Templewhich I thought was fantastic, by the wayand I think its fine for theaters to make use of fun formats like this that cant be replicated at home.
Just know that youre not missing out on all that much if you decide to wait for this movie to come to streaming.
{"blockType":"mv-promo-block","data":{"imageDesktopUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/12\/multicore.png","imageMobileUrl":"https:\/\/images.fastcompany.com\/image\/upload\/f_webp,q_auto,c_fit\/wp-cms-2\/2025\/12\/multicore-mobile.png","eyebrow":"","headline":"\u003Cstrong\u003E\u003Cstrong\u003ESubscribe to Multicore\u003C\/strong\u003E\u003C\/strong\u003E","dek":"Multicore is about technology hardware and design. It\u0027s written from Tokyo by Sam Byford. To learn more visit \u003Ca href=\u0022https:\/\/www.multicore.blog\/\u0022\u003Emulticore.blog\u003C\/a\u003E","subhed":"","description":"","ctaText":"SIGN UP","ctaUrl":"https:\/\/www.multicore.blog\/","theme":{"bg":"#f5f5f5","text":"#000000","eyebrow":"#9aa2aa","subhed":"#ffffff","buttonBg":"#000000","buttonHoverBg":"#3b3f46","buttonText":"#ffffff"},"imageDesktopId":91454027,"imageobileId":91454030,"shareable":false,"slug":""}}
We’re witnessing an unprecedented explosion in creative capability. Voice interfaces are removing barriers for billions who found keyboards cumbersome. AI image generators can mock up virtually any creative direction instantly. The technical constraints that once defined creative work are dissolving.
Yet this abundance creates a new challenge: when everything becomes possible, the possibilities overwhelm us. What then becomes most valuable is knowing whats worth making.
I predict that in 2026, the question “should we build this?” will matter more than “can we build this?”
The capability surplus
The AI conversation is all about capabilities. What you can make. How fast you can make it. What’s now possible.
But there’s a gap emerging between what we can create and what we should create. McKinsey’s November 2025 State of AI report reveals a telling paradox: 88% of organizations now use AI in at least one business function, yet only 39% report enterprise-level financial impact. Theyre capturing value in isolated use cases but struggling to translate that into long-term growth or improved profit margins.
The gap is knowing where to apply it and how to create a framework so that it can actually make an impact.
The skills everyone can hone in 2026
This shift creates genuine opportunity for every creator, professional, and anyone who cares about honing their craft while scaling their impact. When creative execution becomes universally available, three things become differentiators:
Starting with better questions: “How can we have the greatest impact? Which decisions should stay human? Where does automation create fragility?” These aren’t constraints. They’re the frameworks that prevent cognitive overload when everything is technically possible.
Developing taste through iteration: Just as calculators didn’t eliminate the need for mathematical understanding, AI doesn’t eliminate the need for creative foundations. But here’s what changes: the ability to rapidly iterate with AI actually accelerates taste development. You get more attempts, tighter feedback cycles, and faster learning. You build judgment by making more decisions, not fewer.
Knowing when to publish: When AI can generate countless variations instantly, pressing the button to share something with someone becomes the defining creative act. What you send, when you send it, who receives it. These decisions shape identity and message in ways that generation alone cannot.
What tools and platforms can enable now
If knowing what to make is the new skill, the tools that help us develop that skill won’t be just an obsequious yes-man. The most valuable AI tools won’t be those that simply execute your vision, but those that act as creative partners. I predict that tools will emerge that provide the right amount of friction to push your creative ideas.
The role of creative platforms will shift from providing capability to providing capability plus judgment scaffolding built into the product. This means:
Tools that challenge ideas rather than just execute them
Interfaces that know when to stay silent rather than interrupt constantly (fewer notifications, fewer decisions, fewer interruptions)
Features that help users understand why a choice works, not just that it does
The new creative spectrum
We’re moving toward multiple valid modes of creation: human-only, AI-only, AI + human (sometimes openly disclosed, sometimes invisible). Rather than one approach dominating, this spectrum will generate different types of work and different conversations about craft. We will see “not made with AI” declarations coexist with behind-the-scenes AI integration as standard practice.
This reflects an expansion of possibilities. More people will have access to creative tools than ever before. The question is whether they’ll develop the judgment to use them well.
What success looks like now
The optimistic case for 2026 isn’t that AI makes creativity effortless. It’s that AI makes creativity accessible, then rewards those who develop judgment within that access.
Billions of people now have access to professional-grade creative tools. Will we drown in celebrity deepfakes, or will we see an emerging class of contemporary artists? This depends on how well we build “judgment frameworks” into the AI tools we use and our ways of working. We need to use AI tools with discernment, but we also need to hold each other accountable to think deeply and think before we publish.
The most in-demand professionals will be those who can reframe messy questions, challenge false assumptions, and decide what not to optimize. Why? Because when everyone has access to the same generation tools, the baseline quality of output rises, but so does the volume of mediocre work that looks professional but lacks strategic intent.
We’re already seeing the consequences of capability without discernment: marketing campaigns that are technically polished but strategically incoherent, designs that follow trends without serving user needs, code that runs but creates technical debt.
Coca-Cola’s 2024 AI-generated holiday campaign was technically polished but felt “soulless” to audiences who expected the brand’s traditional warmth, while McDonald’s’ Netherlands’ AI holiday ad was pulld after just three days following intense backlash. And in code, GitClear’s 2024 analysis of 211 million lines found that copy-pasted code blocks increased eightfold, generating code that runs but creates the kind of technical debt that compounds into future headaches.
The winners in this new landscapeboth creators and platformswill be those who can cut through the noise. Who develop the human skill to know which problems are worth solving. Who understand that unlimited possibility doesn’t mean every possibility is valuable. The competitive advantage shifts from I can make this to I know this is worth making.
