As companies adopt AI, the conversation is shifting from the promise of productivity to concerns about AIs impact on wellbeing.
Business leaders cant ignore the warning signs. The mental health crisis isnt new, but AI is changing how we must address it. More than 1 billion people experience mental health conditions. Burnout is rising. And more people are turning to AI for support without the expertise of trained therapists. What starts as empathy on demand could accelerate loneliness. Whats more, Stanford research found that these tools could introduce biases and failures that could result in dangerous consequences.
With the right leadership, AI can usher in a human renaissance: simplifying complex challenges, freeing up capacity, and sparking creativity. But optimism alone isnt a strategy. Thats why responsible AI adoption is a business imperative, especially for companies building the technology. That work is not easy, but its necessary.
UNCLEAR EXPECTATIONS
Weve seen what happens when powerful platforms are built without the right guardrails: Algorithms can fuel outrage, deepen disconnection, and undermine trust. If we deploy AI without grounding it in values, ethics, and governancedesigning the future without prioritizing wellbeingwe risk losing the trust and energy of the very people who would lead the renaissance.
Ive seen this dynamic up close. In conversations with business and HR leaders, and through my work on the board of Project Healthy Minds, the signals are clear: People are struggling with unclear expectations around AI use, job insecurity, loneliness, uncertainty, and exhaustion.
In a recent conversation with Phil Schermer, founder and CEO of Project Health Minds, he told me, Theres a reason why professional sports teams and hedge funds alike are investing in mental health programs for their teams that enable them to operate at the highest level. Companies that invest in improving the mental health of their workforce see higher levels of productivity, innovation, and retention of high performers.
5 WAYS TO BUILD AN AI-FIRST WORKPLACE THAT PROTECTS WELLBEING
Wellbeing should be at the core of the AI enablement strategy. Here are five ways to incorporate it.
1. Set clear expectations
Employees need to understand how to work with AI and that their leaders have their back. That means prioritizing governance and encouraging experimentation within safe, ethical guardrails. Good governance builds trust, and trust is the foundation of any successful transformation.
Investing in learning and growth sends a powerful message to employees: You belong in the future were building if youre willing to adapt. We prioritize skill building through ServiceNow University so every employee feels confident working with AI day-to-day.
In a conversation with Open Machine CEO and AI advisor Allie K. Miller, she told me that we need to redefine success in jobs by an employees output, value, and quality as they work with AI agents. This means looking at things like business impact and creativity, not just processes or tasks completed.
2. Model healthy AI behavior
AI implementation is a cultural shift. If we want employees to trust the technology, they need to see leaders and managers do the same.
That modeling starts with curiosity. Employees dont need to be AI experts from day one, but they need to show a willingness to learn. Set norms around when, why, and how often teams engage with AI tools. Ask questions, share experiments, and celebrate use cases where AI saved time or sparked creativity. AI shouldnt be an opt in for teamsit should be part of how we work, learn, and grow. When leaders use AI thoughtfully, employees are more likely to follow suit.
3. Pulse-check employee sentiment consistently
To design meaningful wellbeing programs, leaders must ground analysis in data, continuously improve, and build for scale. That starts by surveying employees to track sentiment, trust, and AI-related fatigue in real time.
Then comes the harder part: acting on the data to show employees theyre seen and supported. Leaders should ask:
Are we tailoring wellbeing strategies to the unique needs of teams, regions, and roles?
Are we embedding empathy into our platforms, workflows, and automated tasks?
Are our AI tools safe, unbiased, and aligned to our values?
Are we making mental health a routine part of manager check-ins?
According to Schermer, The organizations making the biggest strides are the ones treating wellbeing data like commercial data: measured frequently, acted on quickly, and tied directly to outcomes.
4. Focus on connection, keeping people at the center
AI should not replace professional mental healthcare or real-world connections. We must resist the urge to scale empathy through bots alone. The unique human ability to notice distress, empathize, and escalate is largely irreplaceable. Thats why leaders should advocate for human-first escalation ladders and align their policies to the World Health Organizations guidance on AI for health. Some researchers are exploring traffic light systems to flag when AI tools for mental health might cross ethical or personal boundaries.
AI adoption is a human shift, so people leaders need to take responsibility for AI transformation. Thats why my chief people officer role at ServiceNow evolved to include chief AI enablement officer. Todays leadership imperatives include reducing the stigma around mental health, building confidence in AI systems, creating space for open human connection, and encouraging dialogue about digital anxiety, loneliness, or job insecurity.
5. Champion cross-sector collaboration
We need collaboration across industries and leadership rolesfrom tech to healthcare, from HR professionals to policymakersto create systems of care alongside AI. The most effective strategies come from collective action.
Thats why leaders should partner with coalitions to scale access to care, expand AI literacy, and advocate for mental health in theworkforce. These partnerships can help us shape a better future for our people.
THE BOTTOM LINE: AI MUST BE BUILT TO WORK FOR PEOPLE
The future of work should be defined by trust, transparency, and humanity. This is our moment to lead with empathy, design with purpose, and build AI that works for people, not just productivity.
Jacqui Canney is chief people and AI enablement officer at ServiceNow.
Most of the software that truly moves the world doesn’t demand our attention: It quietly removes friction and gets out of the way. You only notice it when it’s broken. That’s not a bug in the business model; it’s a feature. In fact, “unnoticed but indispensable” is the highest customer-satisfaction score you can get.
Consider these categories that already figured this out.
The log-in that isn’t a task anymore
Password managers, once you build the habit, fade into the background. They fill the box before you even remember there was a box. Single sign-on (SSO) systems go a step further and make logging in to everything feel like one action instead of 17 small, annoying ones. And passkeys get rid of passwords entirely. The pattern is consistent: Tools that turn a chore into a non-event ultimately win.
It’s tempting to treat authentication like a “moment”: a page, a button, a ritual. The better approach is to treat it like plumbing. You notice good plumbing by its absence. Otherwise, you just enjoy the hot shower.
Invisible infrastructure already won the internet
Some technologies graduate from “choice” to “ambient.” Transport layer security (TLS) and HTTPS used to be optional. Now they’re table stakes, largely thanks to Let’s Encrypt making it approachable. Your browser nudges everyone toward secure defaults and the ecosystem complies. We don’t “do” TLS; we benefit from it.
This wasn’t always so seamless. In Windows early days, you literally had to install a Winsock stack just to speak TCP/IP. Today, the network stack is simply present, like oxygen. Progress in software often looks like this: The thing we once had to fiddle with becomes the thing we don’t think about anymore.
AI’s next act: not a chat box
Chatbots are neat, but they aren’t the end state of AI. They’re a first draft, like when we used to watch early web pages load images line by line. The real value emerges when intelligent assistance is in the room where work already happens, and it becomes part of the workflow.
In a CRM, the note writes itself while you talk and is already tagged correctly when you hang up.
In design tools, the spec is updated everywhere when you change a component once.
In code review, a suggestion appears inline with a one-click fix, not in a separate AI tab that hijacks your focus.
This is the same story as passwords, SSO, and HTTPS: The win comes from disappearing the steps, not adding a new surface area for attention.
(The funny thing is, most of the work of making AI invisible is just plain old engineering. Yes, there’s lots of AI engineering to make the bots work at all. But plugging them into things in a way that works, that’s the part we’re really behind on.)
BORING ON PURPOSE IS A STRATEGY
At my company we talk about being boring in a specific way: Security and connectivity should feel like electricity. You flip the switch, the lights come on, and nobody argues about the generator or the continent-wide high-voltage distribution network. Being invisible is not the same as being trivial; it’s the reward for sweating details users never see.
Here are five design principles for making software people won’t notice
1. Make the default the decision.
Someone once told me the golden rule of user interface design: If there’s a popup with two options, imagine one of them is “work” and the other one is “don’t work.” Then make “work” the default and delete the popup.
Most users will never visit settings. If the secure, performant, accessible path is the default, adoption happens for free.
2. Budget for latency like it’s a feature.
Under ~100ms, interactions feel instantaneous. Over ~1s, they feel like work. Invisible software feels fast because it never gives the user time to switch contexts. Cache, prefetch, and defer like your product’s life depends on it. Because it does!
3. Automate the paperwork, keep the signatures.
Autofill, SSO, and passkeys are all versions of the same idea: The system should carry the burden. Let humans make approvals and set intent; let machines do the form filling and compliance trail.
4. Progressive disclosure beats feature sprawl.
Hide power tools until they’re needed. The user who needs advanced controls will find them; the one who doesn’t should never meet them. UIs that start simple and get deep on demand feel “light” and earn trust.
5. Fail quietly, recover loudly.
When background systems hiccup, self-heal first. If you must involve the user, say exactly what to do in one step and show you’ve already done the other three. Invisible products don’t turn every exception into a ticket.
THE BUSINESS CASE FOR BEING FORGETTABLE
“Unobtrusive” can sound like “unmonetizable,” but it’s the opposite. Products that vanish into the workflow produce fewer support tickets, shorter onboarding, and more expansion inside organizations. They spread by word-of-mouth because they don’t create new habits; they remove old pain. You don’t need a big campaign to sell relief.
The tricky part is cultural, not technical. Teams must be okay shipping value that isn’t screenshot-worthy. That means investing in the edges: reliability, identity, zero-touch setup, and instant rollback-so customers never have to learn those words.
A SIMPLE TEST
If turning your product off causes immediate, confused swearing from the people who didn’t even know they depended on it, congratulations: you’ve built something great. Now make it a little faster and a little quieter, and do that every quarter.
Because the best compliment your software will ever get is silence.Avery Pennarun is CEO and cofounder at Tailscale.
The trajectory of our national economy is a central concern of every American. Our living costs rise as would-be hegemons battle over neocolonial control through tariff policies. And while social media creativity holds our attention, some part of us recalls older ways of storytelling, and we wonder, where do we belong? Most of us, even newcomers to this countryespecially newcomerswere taught from an early age that anyone who works hard will eventually thrive. But we repeatedly see and know that this is merely a story told to us, not reality.
The community in which you are born has a tremendous impact on your eventual life outcomes. If you are born into a poor community, you will likely remain poor. If you are born into a wealthy one, you are likely to remain wealthy. Author Isabel Wilkerson and socioeconomic researcher Raj Chetty both describe this grating reality. We want to believe in the American Dream, but our eyes see, our ears hear, and our cortisol levels reflect the stress we feel as we strive to reconcile reality with the conflicting narratives of America as a place where anyone can thrive through hard work.
Instead, it is time for a new narrative.
THE POWER OF NARRATIVE IN SHAPING ECONOMIC REALITY
Narrative, more than facts alone, shapes perceptions about who deserves opportunity and resources. Media, pop culture, and policy discourse reinforce or challenge our status quo by elevating the stories of the bootstrapping successful entrepreneur while ignoring stories of the barriers still in place.
After the murder of George Floyd, local TV and the culture turned its attention to topics of structural racism. What followed? Increased business attention on audiences, stakeholders, and customers who were concerned with undoing generations of discrimination. No one with any knowledge of history expected such attention and focus to be permanent. Like looking into the sun, we knew America would quickly avert its eyes. Yet we still hoped that this solar moment would have greater public resonance.
Despite the very public backlash against all things equity, support for diversity, equity, and inclusion persists among many Americans who have experienced the richness and benefit of desegregated life. We now struggle to find the safest words and phrases to describe our internal sense of sharing humanity with otherseven those beyond recently erected walls. This unlabeled value is the seed of a new national narrative.
THE RIGHT TO THRIVE
At Living Cities, we believe the conversation around opportunity must shift from scarcity and survival to abundance and flourishing. When we reframe narratives to center the right of every person to truly thrive, particularly those from marginalized communities, we unlock powerful new possibilities for individuals, families, and entire cities. This positive focus moves beyond merely surviving in systems that were not designed for everyone, toward actively building systems that empower all to grow, innovate, and lead.
By emphasizing narratives of thriving, we foster hope, agency, and dignity. We see entrepreneurs of color not as risky bets but as vital engines of economic growth rooted in resilience and innovation. We recognize neighborhoods historically denied capital not as liabilities, but as sites brimming with untapped potential. This new storytelling affirms that systemic barriers can and must be dismantled, and that access to resources drives shared prosperity, stronger communities, and sustainable development.
Living Cities experience with cross-sector coalitions in cities has shown that using positive narratives of abundance can help community leaders see all individuals as worthy of investment. This helps strengthen community trust, catalyze authentic partnerships, and accelerate economic opportunity. Thriving is more than an aspirational goalit is a proven strategy for revitalizing cities and fundamental motivation for transforming lives.
REFRAME THE CONVERSATION
Living Cities supported city coalitions to use narrative change for direct results. For example, in Albuquerque and Memphis, positive use of narrative enabled loan underwriters to re-examine their assessment of risk related to Black and Latino entrepreneurs.
To reframe the national conversation, organizations and companies can use these best practices in narrative and communications strategies:
Cocreate stories with those affected: Community-led storytelling creates authenticity and greater impact.
Blend hard data with lived experience: Combining human stories with local economic data persuades both hearts and minds.
Invest in media literacy: Teaching audiences to identify and question stereotypes can reduce bias.
Counter negative narratives with abundance, agency, and equity: Highlight systemic successessuch as new Black-owned businesses or increases in affordable homeownershipover deficit-based stories.
INSPIRE A CULTURE OF ABUNDANCE AND EQUITY
Reframing risk as a function of structural barriers, not personal failure, will give us the foundation we need for increased economic opportunity. Storytelling can shift public policy, local business investment, and economic outcomes. Anything is possible when we eliminate our outdated stereotypes and create a new foundation.
Leaders, policymakers, businesses, and media must invest in narrative work as a core equity strategy, reframing the conversation to foster true abundance and agency in Americas communities.
Joe Scantlebury is president and CEO of Living Cities.
Ive spent much of my career in fintech, but some of the most inspiring innovations Ive seen came from a town most people have never heard of.
In early 2025, Ipava State Bank, a tiny community institution in western Illinois, embedded a small amount of life protection into every eligible checking and savings account. No app to install, no portals, no extra stepscoverage was calculated from balances and capped per account. Six months in, reported results included $3.45 million in protection delivered, 7% deposit growth, 4.8% higher average balances, and a 25% increase in customers reaching maximum coverage levelsat a time when many peers were losing deposits.
The program, developed in partnership with Wysh, is part of a growing wave of fintech innovation thats meeting people where they already areat their local banks and credit unions. For The National Alliance for Financial Literacy and Inclusion (NAFLI), its exactly the kind of progress we champion: Technology designed not just for scale, but for inclusion.
Lets talk about why it workedand how other banks could adapt the idea without copying the setting. We worked alongside partners on this effort; here are five observations weve made about the projects design choices any institution can adopt.
1. Default-on beats opt-in.
People dont lack interest in protection; they lack bandwidth. Making the benefit automatic eliminated friction and avoided the shame tax of apply if you can navigate the process. In low-adoption markets, behavioral simplicity is a strategy, not a shortcut.
2. Lead with the institutions trust, not the partners tech.
The coverage showed up through the bank customers already relied on, which reframed the offer from a new product to learn to my bank is taking care of me. Community banks have a trust surplususing it thoughtfully matters more than adding another feature tile.
3. Translate the benefit to local risks.
In Ipava, protection wasnt a perk; it mapped to single-income households, inherited farm debt, and small-business succession. Wherever you operate, write the value statement in the communitys language first, product language second.
4. Measure outcomes the customer can feel.
Deposit growth is great; confidence is the point. Track balance stability, dormant-to-active reactivation, and share-of-wallet movements following benefit awarenesssignals that the relationships getting stronger, not just more expensive to promote.
5. Make branches the on-ramp, not the afterthought.
Frontline staff need a 10-second script. For example, This account now includes a small layer of protectionautomatically and a two-minute FAQ guide. When the explanation is simple, you dont need an app demo to earn adoption.
WHAT THIS CHANGES ABOUT FINANCIAL WELLNESS
Most wellness programs ask people to learn more and do moredownload the app, change the habit, attend the webinar. The Ipava example flips that script: Make the institution do more so the customer doesnt have to.
When protection is embedded where money already lives, inclusion stops being an aspiration and becomes the default state of the relationship. Thats the shift Wysh is helping banks unlockand the kind of design NAFLI believes can redefine what financial literacy looks like in practice.
If your bank is ready to make this shift too:
Dont over-engineer choice. In high-emotion categories, asking users to select multiple options underperform simple and common defaults. If possible, offer clarity, not a catalog.
Dont outsource the story. Tech partners enable; the bank narrates. If customers dont hear it from you, they wont feel it from you.
Dont chase app adoption as the goal. Adoption of the benefit matters more than adoption of the interface. Design to be understood in a branch foyer, not just a home screen.
THE BIGGER INVITATION
If community institutions want to win back deposits and relevance, they dont need shinier featuresthey need more visible care. The lesson from a small bank in western Illinois isnt that every place is Ipava. Its that trust-first, default-on design can work anywhere people still value a bank that shows up for their best daysand their worst.
Maybe the bigger takeaway is simpler: innovation doesnt always look like new technology. Sometimes it looks like a familiar bank doing something timelessshowing up for people when it matters most. And thats why NAFLI is watching this movement closelybecause when fintech starts working for the people who dont download fintech, were finally getting somewhere.
Edwin Endlich is the president and board chairman of The National Alliance For Financial Literacy and Inclusion.
All eyes were on Nvidias quarterly earnings announcement on Wednesday, as investors looked for signs of weakness indicating that the so-called AI bubble is about to deflate. In fact, Nvidia appears to be selling graphics processing unit (GPU) chips for data centers as fast as it can make them.
On the call with analysts, Nvidia reported better-than-expected revenues of $57 billion for its October-ending quarter, a 62% increase over the same quarter last year. Revenues rose by $10 billion, or 22%, from the prior quarter. Perhaps most importantly, the company projected revenues of $65 billion in the current quarter.
As a result, Nvidia shares rose 5% after the earnings were announced at market close on Wednesday. That bump created an additional $205 billion of market capitalization.
Theres been a lot of talk about an AI bubble, Nvidia CEO Jensen Huang said in his opening comments on Wednesday. But from our vantage point, were seeing something very different.
The bubble refers to the possibility that the stock prices and valuations of artificial intelligence companies have become disconnected from their earning potential. Investors also fear that the massive investments that Big Tech and AI companies are sinking into infrastructure like data centers wont be backed up by rapid AI adoption.
Let me remind you that Nvidia is unlike any other accelerator companywe address every phase of AI, Huang said.
Then he set out to show Nvidias current business within the context of some broad technological transitions that he says are happening all at once.
Huang explained that business software that has traditionally run on CPUs is increasingly starting to run on accelerators, specifically the GPUs that Nvidia sells. He said many traditional business tasks are being done by generative AI systems, replacing classical machine learning for things like content suggestion, ad placement, and content moderation.
He also said autonomous AI (such as self-driving cars) and AI agents (such as coding assistants) mark the beginning of yet another big transition: The transition to agentic AI is giving rise to new companies, new products, and new services.
Our singular architecture enables all three of these transitionsacross all industries and all phases of AI, from cloud to enterprise to robots, Huang continued. In other words, Nvidia is set to ride these big waves to big-time chip sales well into the future. Worrying about a bubble today, he seemed to suggest, may be a little short-sighted.
Company CFO Colette Kress said earlier in the call that both hyperscalers like Meta and Google, and top AI labs like OpenAI and Anthropic, continue to spend big on Nvidia chips. We are preparing for aggressive growth ahead and feel optimistic about our opportunity set, she said.
In planning meetings, in brainstorms, in the messy moments when decisions need to be made before all the information is in, AI is my copilot. But not in the cute robot helper way. I treat it like my sharpest strategist, fastest researcher, and most unflinching truth-teller.
As the CEO of Quantious, a future-forward marketing agency that works with tech companies, my job is to stay fast, smart, and endlessly curious; not just for myself, but for my clients. Having executive-level AI by my side is how I operate at scale without sacrificing strategy or soul.
Forget about the hype of AI. Lets talk about what it really takes to work smarter, experiment faster, and free up time to be a creative leadersomething that you cannot automate.
1. AI is my executive sparring partner
When youre running a fast-growing company, youre constantly making judgement calls without all of the details. Most people want ChatGPT to flatter them. I want it to challenge me.
I run new product ideas, positioning statements, and brand hypotheses through AI to surface the cracks I didnt see. I use it to model outcomes, debate assumptions, and yes, poke holes in the perfect plan I thought I had.
Your team might be too polite to challenge you. AI wont be, if you train it well. Start every session with a persona, such as: You are my chief strategy officer. Your job is to challenge mediocrity and raise red flags. Train it over time by giving feedback: Thats too agreeable. Give me a sharper POV or This sounds like fluff. Get specific. And really push it to dig deeper instead of giving you a standard response: This idea solves the problem, but I dont think its the best solution. Push me toward something bolder or more efficient. How would someone with 10x my time/resources/experience approach it differently? You may be surprised where this back-and-forth can take you.
2. I use AI to protect my most valuable asset: Strategic attention
The less time I spend on routine admin tasks, the more time I have to steer the ship. AI is my secret weapon for clearing out the clutter. I use Bluedot to record and transcribe meetingssaving me and my team hours in cleaning up and consolidating notes, and turning around recaps and next steps in minutes. And if I need a detail from the discussion, I can even query the transcript to get the info I need, and all the context around it.
To start using AI for attention management, begin with one task you do often (summarizing docs, doing premeeting research, writing recap emails) and let AI take a pass. If you want to think strategically, you need space to think. AI gives it to you.
3. I never miss a market beat
I don’t have time to read every analyst report or listen to every podcast (who does?!) but I need those insights. AI curates the signal from the noise.
Perplexity Deep Research turns complex trend reports into briefs to share with my team, or even my clients. Waldo gives me market snapshots faster than a team of analysts. Ive also dabbled in AI-powered podcasts, which summarize the most important industry news so I can catch up while on the go. They supplement my other favorite podcasts, so Im always armed with the latest trends and biggest industry moves.
4. I baked AI into the org chart
At Quantious, AI isnt a department. Its a utility, like Wi-Fi or electricity.
Every team has access to tools like ChatGPT, Gemini, and Slack AI. Designers use it to explore creative variations. Ops uses it to document processes faster. Marketers draft content 10 times faster. The tech isnt the point. The enablement is. While not every team member taps into these tools on a daily basis, having them in the toolkit keeps the door wide open for experimentation.
Ive said it before: AI has made remote work more productive, seamless, and well-documented. We dont just integrate AI into workflows; we integrate it into our collective intelligence. Because the point isnt to do more faster, it is meant to elevate how we operate, across the board.
Remember, AI isnt the intern. Its your most strategic hire.
The truth is: Your team doesnt need you to be a prompt engineer. They need you to be an AI-literate leader. AI is no longer a tool in your workflow. Its a seat at your table. Treat it like a trusted advisor, and youll make sharper decisions, faster, without sacrificing strategy or soul.
Lisa Larson-Kelley is founder and CEO of Quantious.
Nvidia forecast fourth-quarter revenue above Wall Street estimates on Wednesday, betting on booming demand for its AI chips from cloud providers against the backdrop of widespread concerns of an artificial intelligence bubble.
The results from the AI chip leader mark a defining moment for Wall Street, as global markets looked to the chip designer to determine if investing billions of dollars in AI infrastructure expansion had resulted in towering valuations that potentially outpaced fundamentals.
The world’s most valuable company expects fiscal fourth-quarter sales of $65 billion, plus or minus 2%, compared with analysts’ average estimate of $61.66 billion, according to data compiled by the London Stock Exchange Group (LSEG).
Shares of the AI market bellwether rose over 4% in extended trading. Ahead of the results, doubts had pushed Nvidia shares down nearly 8% in November, after a 1,200% surge in the past three years. The broader market has declined almost 3% this month.
Still, analysts and investors widely expected the underlying demand for AI chips, which has powered Nvidia’s results since ChatGPT’s launch in late 2022, to remain strong.
Nvidia CEO Jensen Huang said last month that the company has $500 billion in bookings for its advanced chips through 2026.
Big Tech, among Nvidia’s largest customers, has doubled down on spending to expand AI data centers and snatch the most advanced, pricey chips as it commits to multibillion, multi-gigawatt build-outs.
Microsoft reported a record capital expenditure of nearly $35 billion for its fiscal first quarter last month, with roughly half of it spent primarily on chips.
Nvidia expects an adjusted gross margin of 75%, plus or minus 50 basis points, in the fourth quarter, compared with the market expectation of 74.5%.
By Arsheeya Bajwa and Stephen Nellis, Reuters
Since beginning his second term in office, President Trump has taken a sledgehammer to climate action.
His administration has made plans to expand offshore oil and gas drilling, canceled billions of dollars in clean energy projects, rolled back tax credits for electric vehicles, pulled the United States out of the Paris climate agreement, released a report that downplays the risks of climate change, and on and on.
Climate experts have been vocal about the fact that Trump is setting back climate action, which puts the entire world at risk.
The U.S. is the second-most polluting country in the world, behind only China. China, however, has been investing heavily in renewable power, and its total greenhouse gas emissions have been dropping as a result.
Now, a new analysis by ProPublica and the Guardian attempts to quantify what that setback could actually look like.
What the analysis found
Trumps anti-climate policies could release so many extra greenhouse gases over the next decade that they could lead to as many as 1.3 million more temperature-related deaths globally, in the 80 years after 2035, the analysis found.
That estimate covers heat-related deaths, minus the fewer deaths that will occur from cold temperatures. Already, heat is the leading cause of all weather-related deaths, and climate change has led to a noticeable uptick in heat related deaths.
In the U.S. alone, heat-related deaths have increased by more than 50% since 2000, according to the Yale School of Public Health.
The 1.3 million excess deaths does not include, the outlets note, the massive number of deaths from climate changes broader impacts, like droughts, floods, diseases, hurricanes, wildfires, and even lower crop yields.
The number is, admittedly, a small figure when compared to the total number of deaths caused by temperatures changing because of climate change.
A 2021 study on the mortality costs of carbon projected that, between 2020 and 2100, the planet will see 83 million temperature-related excess deaths under a business as usual emissions scenario.
The ProPublica/Guardian analysis acknowledges this, but adds that the figure attributed to Trumps policies speaks to the human cost of prioritizing U.S. corporate interests over the lives of people around the globe.
How the research was conducted
To conduct the analysis, the outlets used scientific models to estimate how many additional emissions will be released into the atmosphere because of Trumps policies. They also took into account the mortality cost of carbon metric, which predicts temperature-related deaths from emissions.
In responses to questions from ProPublica and the Guardian, the Environmental Protection Agency (EPA) contested the science underpinning their analysis, dismissing it as moral posturing. It added that the core calculation method ignores the dramatic uncertainties that dominate long-term climate projections.
But climate scientists say the metric is valid, they report.
Prior to Trump, we had the most ambitious climate policy that the U.S. has ever come up withour best effort to date by far of addressing this growing problem, Marshall Burke, an economist at the Doerr School of Sustainability at Stanford University, told the Guardian.
When we roll these things back, he added, it is fundamentally affecting the damages were going to see around the world.”
The gap between the richest and poorest Americans is widening in what Federal Reserve Chairman Jerome Powell has called a “bifurcated economy,” as the cost of living skyrockets from housing to food prices, but wages for most workers remain stagnant. Basically, high-income individuals are doing well, while lower-income consumers are struggling more and more.
That situation has sparked discussions about whether we’re in a so-called “K-shaped economy.”
A K-shaped economycoined after the shape of the letter: a horizontal line marked by two lines, with one going down and the other uphappens when the economy is rolling along, and then it suddenly loses steam and begins to drop. And then, after a period, the Fed comes in and lowers interest rates to get things going again, professor Peter Ricchiuti of Tulane University’s A.B. Freeman School of Business tells Fast Company.
Simply put, in a K-shaped economy, the Federal Reserve sees the economy weakening, possibly leading to a recession, so it lowers interest rates to stimulate the economy in order to try and avoid that.
“This action really benefits the upper class, as it makes the value of their investmentsstocks, bonds, and real estaterise,” Ricchiuti explains. “More often than not, the wealthy are better off than when the downturn began.”
“Meanwhile, the middle class is hurt even more,” he continues. “If they have any savings at all, its invested in money market funds and bank CDs. These now offer lower returns because interest rates on those instruments have been lowered.”
But “it’s not the Feds fault,” Ricchiuti adds. “The most powerful tool in [the Federal Reserve’s] toolbox is lowering interest rates. Theyre trying to boost the economy but, in doing so, they are widening the economic gap.”
So, are we headed toward a recession?
“I do think the economy is slowing down and potentially moving into recessionary conditions that may show up next year,” Melina Murren Vosse, assistant professor of finance at the University of San Diego’s Knauss School of Business, tells Fast Company. “Talk of the AI bubble, general overvaluations, and global trade uncertainty seem to be making markets squeamish lately.”
Ricchiuti says it’s “tricky” to tell whether we’re heading to a recession “because unemployment numbers are the key indicator of a recession, and we haven’t gotten unemployment levels for quite some time.”
“There just isn’t enough information to feel really comfortable making a determination,” he adds.
That’s in part because the Trump administration fired the head of the Bureau of Labor Statistics (BLS), which collects, crunches, and publishes those unemployment numbers. On August 1, President Donald Trump ordered the firing of Erika McEntarfer after the agency released a report that showed hiring had slowed down significantly over the past three months. Then, a government shutdown further delayed the collection and release of the numbers.
The BLS last released unemployment numbers for the month of August. We are still waiting on September and October numbers, and the BLS said it will not release a full U.S. jobs report for October until it has a full report for November, which it also pushed back to December 16.
Generally speaking, a recession is when there are two consecutive quarters of negative gross domestic product (GDP) growth. But it’s impossible to determine if that’s happened because the numbers haven’t come out.
However, Ricchiuti notes that even though people fear a recession, it generally lasts only a year, while an economic expansion lasts seven years, he says. So even if you’re fearing a recession, it may be more temporary than it might seem.
The Labor Department said Wednesday that it will not be releasing a full jobs report for October because the 43-day federal government shutdown meant it couldn’t calculate the unemployment rate and some other key numbers.
Instead, it will release some of the October jobs data most importantly the number of jobs that employers created last month along with the full November jobs report, now due a couple of weeks late on Dec. 16.
The department’s “employment situation” report usually comes out the first Friday of the month. But the government shutdown disrupted data collection and delayed the release of the reports. For example, the September jobs report, now coming out Thursday, was originally due Oct. 3.
The monthly jobs report consists of two parts: a survey of households that is used to determine the unemployment rate, among other things; and the “establishment” survey of companies, nonprofits and government agencies that is used to track job creation, wages and other measurements of labor market health.
The Labor Department said Wednesday that the household survey for October could not be conducted because of the shutdown and could not be done retroactively. But it was able to collect the hiring numbers from employers, and those will come out with the full November report.
Wednesday’s announcement means the September jobs numbers will likely get extra scrutiny Thursday. They are the last full measurement of hiring and unemployment that Federal Reserve policymakers will see before they meet Dec. 9-10 and decide whether to cut their benchmark interest rate for the third time this year.
The jobs numbers have lately been contentious. After the July jobs report proved disappointing, President Donald Trump abruptly fired the official responsible for collecting the data, Bureau of Labor Statistics commissioner Erika McEntarfer.
McEntarfer herself was quick to say there was nothing suspicious about Wednesday’s announcement. No conspiracy here, folks, she posted on the social media site Bluesky. “BLS was entirely shutdown for six weeks. Payroll data from firms can be retroactively collected for October. The household survey cannot be conducted retrospectively. This is just a straightforward consequence of having all field staff furloughed for over a month.”
____
This story has been corrected to show that the September jobs report is coming out Thursday, not Friday.
Paul Wiseman, AP economics writer
AP Economics Writer Christopher Rugaber contributed to this report.
