Illustration by Tag Hartman-Simkins / Futurism. Source: Getty Images
No matter how much is said about the hundreds of billions of dollars hovering over the tech industry’s head like an anvil in the “AI bubble,” it just won’t pop. Not only is it refusing to budge, but it’s growing by the day.
The AI bubble is getting so bad, in fact, that it’s making previous market bubbles look like chump change.
According to a new assessment by Julien Garran, a research analyst with the firm MacroStrategy Partnership, the AI bubble is now a staggering 17 times the size of the infamous dot-com bubble, a first-of-its-kind run on tech stocks tied to investor hype over the internet. Even worse, Garran estimates that AI now accounts for over four times the wealth trapped in the 2008 subprime mortgage bubble, which resulted in years of protracted crisis across the globe.
In the case of the dot-com bubble, according to macroeconomist David Henderson, major economic catastrophe was avoided as the impact of the stock market rush on US GDP growth was minimal. Unfortunately that isn’t the case with AI investment, which now accounts for a massive chunk of our economic growth after years of unfettered hype.
Prior to the 2008 financial crisis, meanwhile, bullish investors fed into a doomed property market created by banks to turn high-risk mortgages into a font of cash. Like those toxic subprime mortgages, AI has demonstrated very little long-term value — at least at this point in its life, Garran notes.
The trouble with AI, he told MarketWatch, is you “can’t create an app with commercial value as it is either generic [as in video games], which won’t sell, or it is regurgitated public domain [as in homework], or it is subject to copyright.”
He adds that it’s also a hard product to market effectively, as one AI startup in New York City is making clear as its subway adverts get covered with hostile graffiti. All the while, Garran says, the cost of AI systems is growing exponentially larger, with rapidly diminishing gains in capability.
It’s a futile exercise to predict what might finally pop the AI bubble, but one thing’s clear: we’re already at a point of no return.
“To find out whether we have hit a wall we have to watch the LLM developers,” the analyst said. “If they release a model that cost 10x more, likely using 20x more compute than the previous one, and it’s not much better than what’s out there, then we’ve hit a wall.”
Absent AI, Garran warns the economy is already slowing to a crawl, and it’s only a matter of time before the explosive growth in the tech sector begins to reverse, as it did during the dot-com bubble.
Given the sheer size of the threat, it seems the best time for the bubble to burst was yesterday. The second best time, perhaps, is right now.
I’m a tech and transit correspondent for Futurism, where my beat includes transportation, infrastructure, and the role of emerging technologies in governance, surveillance, and labor.
Illustration by Tag Hartman-Simkins / Futurism. Source: Scott Olson / Getty Images
Vermont Senator Bernie Sanders has weighed in with a scorching proposal to confrontone of the most existential employment issues of our time: AI automation.
A long-time advocate for labor rights, Sanders recently released a report about the impact AI could have on jobs over the next decade. The worst-case scenario outlined by the paper is startling: that “AI and automation could destroy nearly 100 million US jobs in a decade.”
The core problem, the report found, is that US workers aren’t receiving their fair share of the profits, which have risen drastically in recent years.
“Since 1973, there has been an explosion in technology and a massive increase in worker productivity,” it reads. “But the resulting economic gains have gone almost exclusively to those at the top. While productivity has risen by 150 percent and corporate profits have increased more than 370 percent, real wages have gone down for the average American worker by nearly $30 a week.”
Here, Sanders is pointing to an economic issue known as the productivity-wage gap. Decades before AI was even a glimmer in Elon Musk’s eye, corporations began suppressing worker’s pay, even as productivity grew, allowing them to pocket the difference. As of 2024, the gulf between worker productivity and wages has become so wide that a fair minimum wage is calculated to be around $25 an hour — over three times the current federal minimum of $7.25.
AI and automation, Sanders is worried, has the potential to make that much worse. Corporate execs aren’t exactly shy about it, either, as they brag about workers whose jobs they get to automate and wield AI like a cudgel to push fewer workers to toil harder for less.
Either way, Sanders has a solution: a “robot tax,”levied against large corporations to distribute to workers whose lives are upended by technological automation. As the report describes it, this would function as a “direct excise tax” on the tech itself, ensuring the “wealth created by these technologies are redistributed back to the workers impacted.”
In other words, it’s a variation on the idea of a universal basic income (UBI), meted out to those directly affected by automation.
“If workers are going to be replaced by robots, as will be the case in many industries, we’re going to need to adapt tax and regulatory policies to assure that the change does not simply become an excuse for race-to-the-bottom profiteering by multinational corporations,” Sanders wrote in his 2023 book “It’s OK To Be Angry About Capitalism,” per Business Insider.
Whether or not the disaster scenario comes to pass remains to be seen. There’s plenty of reason to be skeptical — for example, the fact that AI is failing to generate revenue at 95 percent of the firms that roll it out.
On the other hand, Sanders’ proposal might not go far enough; as BI notes, even Bill Gates supports the idea — a cautionary flag that Sanders’ view might be more mainstream than you’d think.
As some commentators argue, without a vastly expanded social safety net, a UBI can only ever be a band-aid fix leading to a new type of state-backed poverty. For UBI to be truly universal, it must be accompanied by worker protections like rent freezes and price controls, to say nothing of democratic control of the workplace in the first place.
I’m a tech and transit correspondent for Futurism, where my beat includes transportation, infrastructure, and the role of emerging technologies in governance, surveillance, and labor.
Illustration by Tag Hartman-Simkins / Futurism. Source: Getty Images
When your boss insists you to start using AI for your job, do you cave to their demands, or resist using the tech?
The reality is that if you’re in a position where you have to ask, you may be doomed either way.
We suggest you learn from the experience of Kevin Cantera, a man from Las Cruces, New Mexico, who “willingly embraced AI for work” after he and his colleagues were encouraged to use ChatGPT, according to a new entry in the Washington Post’s work advice column.
His reward for going along with those demands, after being a faithful servant for 17 years at the edutech company? Getting replaced by a large language model, along with a couple dozen of his coworkers.
That’s, of course, after his boss reassured him that he wouldn’t be replaced with AI.
Deepening the bitter irony, Cantera — a researcher and historian — had actually grown pretty fond of the AI help, telling WaPo that it “was an incredible tool for me as a writer.” He’d become somewhat of an AI whisperer, claiming he gained a mastery over putting in prompts that helped the AI generate accurate, non-hallucinated info, which he always made sure to review, edit, and rewrite, per WaPo.
“I considered the LLM as a collaborator,” Cantera added. “My productivity was off the charts.”
The LLM was a collaborator alright — a collaborator with his boss.
Cantera is certainly not alone. Bosses the world over are firing their grunts and gleefully replacing them with unproven AI agents. Or they’re culling their workforces down to a skeleton crew, and forcing whoever survived the layoffs to pick up the slack with the help of AI tools. Many of them have openly gloated about it.
But often, quickly trying to replace human know-how with constantly hallucinating tech blows up in the faces of the bosses who embraced it, sending them scrambling to rehire humans.
Plus, a bevy of research is being published that pokes gaping holes in the tech’s promises of revolutionizing productivity, like one MIT study that found that an astonishing 95 percent of companies that integrated AI saw no meaningful growth in revenue, and another showing that it created a frustrating and mired workflow in which employees pass down shoddy AI “workslop” that ends up creating more work for their colleagues who need to fix the botched efforts.
With dozens of his colleague now out of the picture, Cantera is wondering if there will even be anyone left to look over the “workslop” at his former employer.
“I am sure they try to vet what they get from AI, but I also know that they let go of a good number of [subject-matter experts],” Cantera told WaPo. “It is absolutely terrifying to think they could be relying on the model’s output without any [quality assurance] or true subject-matter expert review.”
But circling back to the original question, should you assent to or dissent from your boss’s demands to use AI? If they’re probably planning on replacing you anyway, you may as well go out on your own terms.
I’m a tech and science correspondent for Futurism, where I’m particularly interested in astrophysics, the business and ethics of artificial intelligence and automation, and the environment.
Researchers believe they have developed the first dedicated blood test to diagnose chronic fatigue syndrome (CFS), also known as myalgic encephalomyelitis (ME), which has the potential to be a game-changer for millions around the globe who suffer from the debilitating condition.
Researchers from the University of East Anglia (UEA) and Oxford BioDynamics, in collaboration with The London School of Hygiene & Tropical Medicine and Royal Cornwall Hospitals NHS Trust, have announced they’ve successfully developed the first blood test for ME/CFS that delivers 92% accuracy in identifying the condition.This bypasses the current path of diagnosis, where symptoms are evaluated and a host of blood tests are performed to rule out other conditions such as thyroid problems or anemia.
It could also mean validation for people suffering from ME/CFS, a condition that’s often misdiagnosed and remains poorly understood, resulting in people living with it for a significant period of time before they get answers. Treatment is also challenging.
“This is a significant step forward,” said Dmitry Pshezhetskiy, a professor from UEA’s Norwich Medical School. “For the first time, we have a simple blood test that can reliably identify ME/CFS – potentially transforming how we diagnose and manage this complex disease.
“ME/CFS is a serious and often disabling illness characterized by extreme fatigue that is not relieved by rest,” he added. “We know that some patients report being ignored or even told that their illness is ‘all in their head.’ With no definitive tests, many patients have gone undiagnosed or misdiagnosed for years.”
The test was developed using Oxford BioDynamics’ EpiSwitch 3D Genomics technology, which analyzed DNA folding in blood samples taken from 47 patients with severe ME/CFS and 61 individuals without the condition. Each of our cells contains around two meters (6.6 feet) of folded DNA, which determine how genes are switched on and off. While the condition isn’t genetic – we’re not born with genetic mutations for it – in recent years scientists have gained greater knowledge of it having an epigenetic basis. Essentially, external influences change how genes are modulated, and the blood test can identify these markers of ME/CFS.
“Chronic fatigue syndrome is not a genetic disease you’re born with,” said Alexandre Akoulitchev, Chief Scientific Officer at Oxford BioDynamics (OBD). “That’s why using EpiSwitch ‘epigenetic’ markers – which can change during a person’s life, unlike a fixed genetic code – was key to reaching this high level of accuracy. The EpiSwitch platform behind this test, together with OBD’s vast 3D Genomic knowledge-base, has already been proven to deliver practical, rapid blood diagnostics accessible at scale.
“With this breakthrough, we are proud to enable a first-in-class test that can address an unmet need for a quick and reliable diagnostic for a complex, challenging-to-identify illness,” he added.
EpiSwitch technology has already been used in detecting markers in amyotrophic lateral sclerosis (ALS) and rheumatoid arthritis, and a blood test for prostate cancer – EpiSwitch PSE – is already in use, with 94% accuracy.
The researchers claim that the blood test identified 92% of the ME/CFS cases in the analysis cohort, however, some medical experts remain skeptical about whether this diagnostic tool is the breakthrough the team says it is.
“The authors are claiming a higher rate of sensitivity and specificity than in most other biomedical tests,” said Dr Alastair Miller, a retired physician in infectious disease and internal medicine. “My main concern with this study is the lack of appropriate controls. They are using healthy controls rather than those with other chronic conditions such as depression or fibromyalgia or even MS (multiple sclerosis). My worry is that it will prove to be yet another false dawn, launched with a huge amount of hype and will raise patients’ expectations unrealistically.”
Others have also questioned the method of testing for the condition, suggesting it’s too early to take the success at face value.
“Despite the test predicting 22 of 24 cases and 44 of 45 controls in independent samples, these claims are premature,” said Professor Chris Ponting, Chair of Medical Bioinformatics at the University of Edinburgh. “This is because results could be confounded in three ways: by (1) Sex and/or by age: sex- and age-matching was not done (beyond matching age criterion of 20-80 years old); (2) Batch: all cases were from the CureME Biobank, whereas most (41 of 61) controls came from the company’s own biobank; and, (3) Inactivity and severity: all cases had severe symptoms and were house-bound, whereas all controls were healthy and likely physically active.
“This test needs to be fully validated in better designed and independent studies before it is considered for clinical application,” he added. “Even if validated, the test will be expensive.”
Nonetheless, the research itself offers new clues as to the epigenetic triggers of ME/CFS and could also further our knowledge of identifying and treating long COVID.
“Understanding the biological pathways involved in ME/CFS opens the door to developing targeted treatments and identifying which patients might benefit most from specific therapies,” said Pshezhetskiy. “We hope that the Episwitch CFS test could become a vital tool in clinical settings, paving the way for more personalized and effective care.”
The researchers also added that wider database testing will be needed to confirm the findings that emerged from their testing population.
“Larger-scale, prospective validation in multi-centre cohorts will be essential to establish external validity,” they wrote. “The observed epigenetic changes may be downstream of chronic immune activation; longitudinal and interventional studies (e.g., before/after immunomodulatory therapy) are needed to test causality.”
In July, researchers at Jackson Laboratory announced their own diagnostic breakthrough for ME/CFS, with a test that identified the condition through specific immune cell and gut profiles.
Summary: A small peptide called CAQK, composed of just four amino acids, has shown remarkable neuroprotective effects in mouse and pig models of traumatic brain injury. When injected intravenously, CAQK travels directly to damaged brain tissue, where it binds to overexpressed proteins and reduces inflammation, cell death, and tissue damage.
Treated animals displayed improved memory and motor function without toxicity, suggesting strong potential for safe, non-invasive therapy.Researchers plan to seek FDA approval for human clinical trials, marking a major advance toward drug-based treatments for traumatic brain injury.
Key Facts:
Targeted Repair: CAQK homes in on injured brain areas, reducing inflammation and cell death.
Pre-Clinical Success: Improved recovery in mice and pigs without toxicity; human trials planned.
Source: CSIC
Advanced Chemistry of Catalonia (IQAC) of the Spanish National Research Council (CSIC), an institution under the Ministry of Science, Innovation and Universities, has discovered that a small compound—a peptide made up of four amino acids called CAQK—has a significant neuroprotective effect in mouse models of traumatic brain injury.
When administered intravenously shortly after injury in animal models (mice and pigs), CAQK specifically targets the damaged areas of the brain, attracted by a protein that is overexpressed in injured tissue following trauma.
Finding a non-invasive way to treat an injured brain is one of the major challenges in neurology. Credit: Neuroscience News
CAQK accumulates in the region marked by this protein and is able to reduce inflammation, cell death, and damage to brain tissue. Moreover, in mice, it improved functional recovery without apparent toxicity.
The results, published in the journal EMBO Molecular Medicine, open new possibilities for treating injured areas of the brain.
The study was led by the company Aivocode (a spin-off of the Sanford Burnham Prebys Institute) in San Diego, California, in collaboration with the Institute for Advanced Chemistry of Catalonia (IQAC-CSIC) and the University of California, Davis.
Aivocode, founded by researchers Aman P. Mann, Sazid Hussain, and Erkki Ruoslahti (authors of the study), plans to soon seek authorization from the U.S. Food and Drug Administration (FDA) to begin Phase I clinical trials in humans.
Although no specific date has been set, the fact that CAQK is a short peptide—easy to produce and with good tissue penetration—makes it a strong candidate for drug development.
Traumatic Brain Injury
Traumatic brain injury (TBI) is brain damage typically caused by blows to the head, such as those resulting from traffic accidents, workplace incidents, or falls. It is estimated to affect around 200 people per 100,000 inhabitants each year.
Currently, treatment focuses on stabilizing the patient by reducing intracranial pressure and maintaining blood flow, but there are no approved drugs to halt brain damage or its secondary effects, such as inflammation or cell death. In addition, the therapies under investigation require direct injections into the brain, an invasive technique that can cause complications.
“The current interventions for treating acute brain injury aim to stabilize the patient by reducing intracranial pressure and maintaining blood flow, but there are no approved drugs to stop the damage and secondary effects of these injuries,” explains Dr. Pablo Scodeller, researcher at IQAC-CSIC and co-author of the study.
The Great Challenge of Neurology
Finding a non-invasive way to treat an injured brain is one of the major challenges in neurology. This study moves in that direction, building on previous work carried out by the researchers in 2016 and published in Nature Communications.
At that time, researcher Aman P. Mann, together with Pablo Scodeller, working in the laboratory of Dr. Ruoslahti (senior author of both studies) at Sanford Burnham Prebys, discovered a peptide—a small chain of amino acids, the building blocks of proteins—that specifically targeted injured areas of the brain in mice.
The peptide, named CAQK, was identified through a large-scale screening technique known as peptide-phage display, which allows the selection of molecules with affinity for specific tissues. In that earlier study, CAQK was used as a “vehicle” to deliver drugs directly to the damaged area.
However, in their new work, the researchers went a step further and demonstrated that the CAQK peptide itself has therapeutic effects.
To evaluate its therapeutic activity, the peptide was first administered intravenously shortly after a moderate or severe traumatic brain injury, and it was observed that the peptide accumulated in the injured brains of mice and pigs (the latter having brains more similar to humans than mice).
Furthermore, it was found that the peptide binds to special molecules called glycoproteins (proteins attached to sugars), which become more abundant after an injury and are part of the extracellular matrix—a supporting network that surrounds brain cells.
Treatment of mice with traumatic brain injury using this peptide resulted in a reduction in lesion size compared to control mice.
“We observed less cell death and lower expression of inflammatory markers in the injured area, indicating that CAQK alleviated neuroinflammation and its secondary effects. Behavioral and memory tests conducted after treatment also showed improvement in functional deficits, with no evident toxicity,” explains the study’s first author, Dr. Mann.
The study’s results demonstrate that the CAQK peptide can help repair the damaged area, highlighting its potential therapeutic applications following trauma.
“What’s exciting is that, in addition to proving highly effective, it’s a very simple compound—a short peptide that is easy to synthesize safely at large scale. Peptides with these characteristics show good tissue penetration and are non-immunogenic,” concludes Scodeller.
Key Questions Answered:
Q: What makes CAQK different from other brain injury treatments?
A: It can be administered intravenously and selectively targets injured brain tissue without invasive procedures.
Q: How does CAQK work?
A: The peptide binds to specific glycoproteins overexpressed after brain injury, reducing inflammation and protecting neurons.
Q: When could human trials begin for using CAQK to treat TBI?
A: Researchers plan to seek FDA authorization soon to start Phase I clinical testing.
About this neuropharmacology and TBI research news
Author: Pilar Quijada Source: CSIC Contact: Pilar Quijada – CSIC Image: The image is credited to Neuroscience News
The Rundown: We sat down with OpenAI CEO Sam Altman at Dev Day 2025 for a wide-ranging conversation on the company’s new launches, AGI, the future of work, the rise of AI agents, and more.
The details: Altman said AI’s ability for “novel discovery” is starting to happen, with recent scientists across fields using the tool for breakthroughs. Altman thinks the future of work “may look less like work” compared to now, with a fast transition potentially changing the “social contract” around it.He believes Codex is “not far away” from autonomously performing a week of work, saying the progress of agentic time-based tasks has been disorienting.The CEO also highlighted the potential for a zero-person,billion-dollar startup entirely spun up by a prompt being possible in the future with agentic advances.
Why it matters: Dev Day 2025 gave us a new step in both ChatGPT and OpenAI’s agentic tooling evolution, and Altman’s commentary provided an even deeper look into the future the company envisions. But no matter how strange the AI-driven changes get, Altman remains confident in humanity’s ability to adapt and thrive alongside them.
In this explosive interview, Professor Jeffrey Sachs breaks down the escalating tensions in the Middle East, revealing critical developments between Israel and the United States that could reshape global politics. 🌍⚠️ From rising geopolitical risks to behind-the-scenes power plays, Sachs exposes the dangerous path the world may be heading toward. Don’t miss this urgent analysis — it could affect us all. 👉 Watch till the end for eye-opening insights that mainstream media won’t tell you. 🔔 Subscribe for more breaking geopolitical updates, expert analysis, and exclusive interviews.
A clip from Episode 38 breached 4 million views worldwide, here we follow up the focus on migration with a new documentary on how it’s reshaping Germany and Poland. pic.twitter.com/7O12UkkF3V
Anthropic scored its biggest enterprise win yet. On Monday, the company announced a partnership with consulting giant Deloitte to offer its Claude models to the firm’s 470,000 employees.
The partnership expands a deal made between the two companies last year, in which Deloitte planned to train 15,000 professionals on Claude.
Deloitte will establish a “Claude Center of Excellence” as part of the deal, which will include specialists to develop implementation frameworks to scale AI pilots within the company.
Deloitte’s Chief Strategy and Technology Officer Ranjit Bawa stated in a press release that the partnership reflects the companies’ aligned approach to responsible AI.
Terms for the partnership weren’t disclosed, but Paul Smith, Anthropic’s chief commercial officer, told CNBC that “We are both investing a significant amount in this partnership, whether that’s financial or whether it is just simply the engineering resource that we’re going to put into this as well.”
The deal signals a broader push into the enterprise market by major AI firms as they seek to monetize their massive models.
However, enterprises are still largely struggling to derive value from these deployments. An MIT study published in August found that 95% of organizations surveyed have seen no returns on their investments in generative AI, despite $30 to $40 billion being poured into the technology
Image source: The Rundown
