In Central Texas, a bitter fight over a $1 billion water project offers a preview of the future for much of the state as decades of rapid growth push past the local limits of its most vital natural resource.On one side: Georgetown, the fastest growing city in America for three years straight, which in 2023 signed a contract with an investor-funded enterprise to quickly begin importing vast volumes of water from the Simsboro Formation of the Carrizo Wilcox Aquifer, 80 miles to the east.On the other side: the cities atop the Simsboro that rely on its water. Bryan, College Station, and the Texas A&M University System, a metro area with almost 300,000 people, have sued the developer to stop the project. A trial is set for the first week of May.The site of a water pipeline project by the company Recharge through Lee County into Williamson County is pictured on March 28. [Photo: Dylan Baddour/Inside Climate News]Were going to fight this thing until the end, said Bobby Gutierrez, the mayor of Bryan. It effectively drains the water source of the cities.The pump and pipeline project to Georgetown, developed by California-based Upwell Water, is the largest of at least a half dozen similar projects recently completed, under construction or proposed to bring rural Carrizo Wilcox aquifer water into the booming urban corridor that follows Interstate 35 through Central Texas.It would eventually pump up to 89 million gallons per day, three times the usage of the city of Bryan, to Georgetown and its neighboring cities.That basically stops all the economic development we have, Gutierrez said. Were talking about our survival.The fight over the Upwell project could well be a prelude for the broader battles to come as cities across Texas outgrow their water supplies. Lawmakers in the state Capitol are pushing to avert a broad scarcity crisis with funding to desalinate seawater, purify salty groundwater, and treatoilfield wastewaterto add to the supply. But all of these solutions remain years from realization. In the near term, only import projects from freshwater aquifers will continue to meet the growing water demands of thirsty Texas cities.Regulation of such projects falls to a patchwork of small, rural agencies called groundwater conservation districts, which might not be fully equipped or empowered to manage plans for competing regional water needs that can affect entire cities for generations to come.Texas law offers limited clarity, generally preferring a landowners right to pump their own groundwater over regulations on private property. Despite fierce denunciations of the Upwell project from nearby city leaders, no one has alleged that its developers have broken any laws.Were following the rules. Why are we being vilified? said David Lynch, a managing partner at Core Capital investment firm in Houston and a partner in the Upwell project. I think they feel uncomfortable about whats coming and their reaction is to make us go away.After all, hes not the only one doing this. Five years ago, San Antonio started pumping up to 49 million gallons per day through a 140-mile pipeline from the Carrizo Wilcox Aquifer. Another pipeline was completed last year and will soon begin pumping to the city of Taylor and the new Samsung microchip manufacturing complex there. Another, scheduled for completion this year, will take water into the cities of Buda and Kyle.After the lawsuit delayed the Upwell projects tight timeline, Georgetown commissioned two other pipeline projects from the same aquifer.[Image: Paul Horn/Inside Climate News]People are starting to pay enough for water to make these sorts of projects work, Lynch said, driving his black Ford Super Duty Platinum truck down the dirt roads of Upwells 9,000-acre farm property and well field in Robertson County. Theres no cheap water left in Texas.In the middle of all this is the little Brazos Valley Groundwater Conservation District, based in the small town of Hearne and also a defendant, alongside Upwell, in the lawsuit.District manager Alan Day feels for the cities of Bryan and College Station. To an extent, he said, theyre right. The more pumping from the aquifer, the sooner everyone will reach conditions of scarcity, though he doesnt think it will happen as quickly as city leaders say.At the same time, he said, Bryan cant claim the water. Groundwater is a private property right in Texas as sacred as any other. Everyone is allowed to pump whatever their land produces.Water is the new oil, said Day, a former ranch manager of 27 years. They have a commodity that can be sold and they have every right to sell it.At this time, he said, he has no authority to stop landowners from pumping as long as they fulfill the requirements of the permitting process, which Upwell did. Even if he could do it, Day chuckled at the notion that state leaders would let his tiny office put the brakes on development along the I-35 corridor, home to manufacturing campuses of Tesla, Samsung, and Apple, and offices of Amazon, Meta, and Google, as well as one of the nations largest clusters of data centers and its fastest growing cities.However, Day said, there will come a day when that changes. The laws for his district, like all others in Texas, specify a threshold at which new rules kick in. Its called the desired future condition, or DFC, a level below which the district is not willing to go. When they get there, everyone will face restrictions on pumping and the days of groundwater abundance will be over for the Simsboro portion of the aquifer. To date, no district in Texas has hit its DFC.Day said hes only following the rules. Hell honor the property rights of landowners who want to pump, and when they hit the DFC, hell implement restrictions district-wide.What does that do to the growth of Bryan and College Station and Texas A&M and anyone else who is depending on Simsboro? Day asked. It stops it.The offices of the Brazos Valley Groundwater Conservation District in the tiny town of Hearne. [Photo: Dylan Baddour/Inside Climate News]The Texas MiracleThis situation follows a generation of steep growth and development that state leaders have dubbed the Texas Miracle. The population of Williamson County, seated in Georgetown, 28 miles north of Austin, doubled in 17 to 700,000 people while itsmedian household incomeincreased by more than 90%. Neighboring counties share similar stories, where sprawling subdivisions and shimmering tech campuses now cover former ranchlands.Georgetown needs to add millions of gallons per day to its water supply within the next several years. When it signed the pipeline contract in 2023 that stipulated deliveries beginning in 2030, it was acting on a much tighter timeline than decades that are typically considered for large scale water planning.Based on hyper growth that weve seen in our water territory, weve seen the need for higher levels of contracted water sooner than we originally anticipated, said city manager David Morgan.Most of the new water will serve new residential areas, he said, and will be used primarily to irrigate lawns and other neighborhood landscaping. Williamson County is also courtinga cluster of five large data centersthat it expects would bring another 100,000 people to the county.But what if Bryan, and the cities of the Brazos Valley, want data centers, too? The region is currently pursuing ambitious opportunities in semiconductors, nuclear energy, aerospace, defense, and life sciences, said Susan Davenport, president of the Greater Brazos Partnership, an economic development group.These sectors, along with the growing workforce and families who support them, are directly dependent on access to our local water resources, she said.Gold Rush on WaterAlthough many major projects importing groundwater into Central Texas are just now being realized, the plans have been in the works for decades, according to Michelle Gangnes, a retired finance lawyer and co-founder of the Simsboro Aquifer Water Defense Fund.In 1998, Gangnes moved from Austin to rural Lee County. That same year, San Antonio, 140 miles away, announced plans to import 49 million gallons per day from wells in Lee County on the site of an old Alcoa aluminum smelter. A prolonged fight ensued and the project was never realized, but many others would follow.Thats what started the whole gold rush on water, Gangnes said. It resulted in all these groundwater districts being formed, trying to resist the water rush on the Simsboro.The groundwater districts were formed by an act of the Texas legislature in 2001. But, when the time came to make groundwater rules, powerful interests kept them loose, according to Ken Kramer, who previously directed the Texas office of the Sierra Club for 24 years. Chief among them was T. Boone Pickens, the iconic Texas oilman who also wanted to export groundwater from his land holdings in the Panhandle.There was heavy lobbying by groundwater exporters to make sure that groundwater districts could not stop exports, Kramer said. Groundwater then became more of the target for moving water to growing areas and populations.A sign on a water pipeline scar in Lee County on March 28. [Photo: Dylan Baddour/Inside Climate News]Under a principle in Texas called the right of capture, landowners are allowed to pump from their land whatever they are able to. Changes made to the Texas Water Code in 2001 stipulated that withdrawals are allowed so long as they dont affect other permit holders unreasonably, which lacks a firm legal definition. That leaves lots up to interpretation for the groundwater districts of Texas.They live in a difficult world where its unclear exactly what their power is to tell somebody no, said Robert Mace, executive director of the Meadows Center for Water and the Environment at Texas State University. If you tell somebody no youre almost guaranteed to get sued.In recent years, several major pipeline projects into Central Texas came online. San Antonio eventually got its Carrizo Wilcox Aquifer water through a 6-foot-wide, 140-mile-long Vista Ridge pipeline which began drawing water from Burleson County in 2020, causing levels in neighboring landowners wells to plummet.The old Alcoa wells in Burleson County were also put to use. A developer called Xebec Holdings bought the 50-square-mile property in 2022 and signed deals to pipe almost 18 million gallons per day to the City of Tyler.Theres constantly people out there trying to lease water rights to see if they could do a project to sell water, said Gary Westbrook, general manager of the Post Oak Savannah Groundwater Conservation District. Were going to have to find a way to regulate. You cant just say no.The Gatehouse Pipeline is currently under construction to Georgetown, with another one called Recharge in development. Morgan, the Georgetown city manager, said those two projects were identified and accelerated after the lawsuit challenged the Upwell project.We believe the lawsuit is going to likely delay getting that fully resolved, he said.The Upwell ProjectUpwell Water, a San Francisco-based financing firm,announcedin 2020 that it had raised $1 billion from investors to monetize water assets.Upwell partnered with CoreCapital investors in Houston, which bought its 9,000-acre Robertson County farm property in 2021. Lynch, the managing partner at CoreCapital, said he expected to sit on the property for 10 years until the economics of water made it attractive to develop a major export project.But as soon as he entered the market, he found eager buyers willing to pay well.We bought it and all of a sudden we had everybody calling saying we need water, Lynch said. Then we said, we have more demand than we can supply, lets talk to the neighbors.Upwell recruited seven neighboring landowners to put company wells on their property and contribute to the export project.These arent regular irrigation wells, which in this area can tap water 40 feet down. These are 1,400 feet deep, cased in 2-foot-wide steel pipe, able to produce large volumes.Mark Hoelscher, a landowner who is selling groundwater from his land to the Upwell project, stands in front of a 1,200-foot-deep Simsboro Well in rural Robertson County on March 20. [Photo: Dylan Baddour/Inside Climate News]Its a million-dollar hole, said Mark Hoelscher, one of the neighboring landowners involved in the project, as he looked up at one of the diesel-powered well installations. Its big time.In October 2022, Upwellreceived permitsfor 16 wells to pump nearly 45 million gallons per day without any challenges in the hearing process. Four months later it receivedits permitto export the water out-of-district. Then in September 2023, the district issued permits for another 32 wells belonging to the seven adjoining landowners to produce an additional 45 million gallons per day.Until that point, authorities in the Bryan-College Station metro area, some 30 miles south, apparently remained unaware of the project transpiring in Robertson County. Not until September 2024, when the district considered applications for updated permits to export the combined 89 million-gallon-per-day production of all 48 wells, did Texas A&M University enter into the proceedings, filing a request for review by the State Office of Administrative Hearings.Texas A&M University declined to comment for this story.No one has questioned the fact that we own the land and we have rights to the water underneath it, said Hoelscher, a third generation landowner in the Brazos River Valley. The fact of the matter is the water is ours.The LawsuitOne week later, A&M fileda lawsuitin state district court seeking a temporary injunction stopping the groundwater district from recognizing any of the permits associated with the Upwell project until a hearing is held.A&M argued that the previously issued permits should be open for re-examination because some board members of the groundwater district were ineligible for service at the time the permits were originally approved.In November, Bryan and College Stationfiled papersto join the lawsuit. It said their ability to produce groundwater from their Simsboro wells and the economic vitality of the region will be adversely affected if the Contested Applications are granted.College Station Mayor John Nichols, a former professor of agricultural sciences at Texas A&M, said in a statement: The transfer of groundwater from our district to users in other areas is one of the most significant issues facing the College Station/Bryan area. Im a staunch proponent of private property rights, but we are deeply concerned about the long-term impact of excessive extraction on our community.He called on lawmakers to adopt statewide groundwater regulations ensuring the rights of current permit holders over new water users.None of that, however, matters to the trial that will take place in early May. All the judge will decide is whether or not A&M and the cities have rights to challenge the previously issued permits.In court filings, Upwell argued A&Ms petition demands that the Court turn back time and recognize a nonexistent right to administratively contest final groundwater permits that the Brazos Valley Groundwater Conservation District properly noticed and issued to Intervenors months and years priorall without any complaint or contest by any party, including Plaintiff.If the judge denies A&Ms request, the permits will be issued and work will begin on the Upwell project pipeline.If the judge grants A&Ms request, the permits will head into a potentially yearslong process of state administrative hearings that could threaten the viability of the project and its promised returns to investors.Construction on a water tank and tower, part of a Manville Water Supply Corp project through Lee County to Williamson County, on March 28. [Photo: Dylan Baddour/Inside Climate News]Desired Future ConditionWhether or not the pipeline gets built, other similar projects are likely to follow. The situation is headed in one direction: towards the DFC, the threshold at which restrictions begin.In the Brazos Valley and surrounding districts, that threshold is a 262-foot drop in water wells from levels measured in 2000. In the 25 years since then, pumping has led the wells water to drop by one quarter of that allotted reduction, according to district manager Day, suggesting ample water supplies remain.But, that remains to be seen. In total, Day said his district has issued permits for up to 291 million gallons per day of pumping from the Simsboro Formation, averaged yearly, of which 89 million gallons per day are associated with the Upwell project. However, only a fraction of that permitted volume is actually pumped.If all permitted pumping were to suddenly come online, Day said, computer models showed they would hit the DFC in six years.In reality it wont happen quite that fast. The Upwell project plans to scale up its pumping gradually over years. And many farmers hold irrigation permits to pump much more water than they ever actually will, unless they also encounter the opportunity to join an export project.When the aquifer hits the DFC, the rules say it mustnt fall further. That means all users would face mandatory curtailment. Its unclear how such unprecedented measures would be enforced in Texas.For Gutierrez, the mayor of Bryan, this management method creates a contest for investors to tap the water-wealthy Simsboro Formation and sell off its bounty before time runs out.They want to exploit everything we have for their personal benefit, he said. Its a race of who can take the most amount of water in the least amount of time to deplete a resource for their pocketbooks.Dylan Baddour, Inside Climate NewsThis article originally appeared on Inside Climate News. It is republished with permission. Sign up for their newsletter here.

Max just got a new logo. Again. Two years after rebranding from HBO Max to just Max with new a bright blue-and-white logo, the Warner Bros Discovery-owned streaming service is making an update to its logo. This time, its swapping blue for a metallic black and white logo.According to Max, the color change is part of a larger refresh. Max says the standalone logo will be in the black-and-white color scheme, but an updated color palette, chosen to allow for flexibility of the logo in app and in marketing materials, will be unveiled in the coming months.[Images: HBO Max]Why Max updated its logoMax includes content from HBO and other Warner Bros Discovery brands, like Adult Swim, Animal Planet, Cartoon Network, CNN Films Discovery, and TNT, but the new logo appears to put HBOwhich is responsible for top shows for the streamer like The White Lotus, The Sopranos, and Successionback at the center. The new logo reflects the black-and-white color palette of HBOs branding and retains the circle inside the counter of the A in Max, a callback to the circle inside the O in the HBO logo.[Images: HBO Max]Throughout the streaming wars, individual brands have updated their visual identities to stand out in a sea of blue logos. Disney+ updated its logo last year from blue to teal, and when Max first rolled out its blue logo, its former global chief marketing officer Patrizio Spagnoletto said the specific shade was chosen because it stood apart from Paramount blue and Prime blue. Together with the logo mark, the color communicated something about how the streamer wanted to be perceived, he said.With our blue and the way that the logo is designed, what we were going for is a combination of premium but accessible, Spagnoletto said in 2023.In black and white, the new Max logo seems to be amping up the premium aspect of its brand and downplaying the accessible. A Nielsen survey of the top 10 most-streamed shows in the U.S. may suggest why, with HBO shows like True Detective and The White Lotus among the few Max shows with enough viewers to make it onto the list dominated by Netflix, Disney+, and Hulu shows. Its a strategy that just might work.

Imagine walking into a space where ancient echoes meet the dynamism of the present. Thats the vibe Oculta Studio is bringing to Milan Design Week 2025 with their captivating Kinetic Vases. After a debut that sparked curiosity at Mexico City Art Week 2025, these arent your grandmas dusty heirlooms. Theyre a vibrant reimagining of traditional ceramic vases, infused with the unexpected element of movement, turning them into mesmerizing kinetic sculptures.Designer: Oculta StudioWhats so compelling here is how Oculta Studio is playing with our perceptions. Were used to seeing historical vases as static objects, frozen in time. But these pieces challenge that notion head-on. Theyre not just beautiful to look at; they do things. They shift, they transform, prompting us to think about the inherent fragility and constant evolution of the objects that surround us. Its like the past isnt fixed but is instead in a perpetual state of becoming.The studios exploration delves into some pretty profound ideas. By introducing kinetic elements, theyre not just making eye-catching art; theyre making a statement about impermanence. Think about it these are vessels, traditionally meant to hold and preserve. Now, theyre actively changing, almost as if theyre breathing or responding to the passage of time. This controlled transformation, through what the studio describes as controlled fractures and shifting forms, beautifully illustrates the transient nature of both material and memory. What we hold dear, what we remember its all subject to change.Set to be a highlight in Milans vibrant Isola District from April 7th to 13th, the exhibition promises a unique experience. Youll witness firsthand the intriguing intersection of time-honored craftsmanship and cutting-edge kinetic innovation. Its a blurring of boundaries, really are these historical artifacts being reinterpreted, or are they entirely new forms of artistic expression inspired by the past? Perhaps its a bit of both, and thats what makes it so engaging. Oculta Studio isnt just presenting pretty objects; theyre sparking a dialogue. Theyre inviting us to reconsider our relationship with the artifacts of history and to embrace the beauty inherent in change and impermanence. Keep an eye out for these Kinetic Vases theyre poised to be a thought-provoking and visually stunning contribution to Milan Design Week 2025.The post Oculta Studios Kinetic Vases Reimagine Tradition at Milan Design Week 2025 first appeared on Yanko Design.

Ross Lovegrove designed the PolarisGo chair with data from the Polaris Dawn space mission, injecting magnesium into its design. Constructed from aerospace-grade aluminum alloy, Lovegrove collaborated with CreativeWorkStudios, drawing inspiration directly from the missions launch shock wave data. Lovegrove revealed his access to the flight data, explaining his unique approach I have taken this data and translated it into a frequency that is passed through the frame of the GO Chair creating four distinct polarized focus points in each corner of the seat, said the designer.He refers to the chairs main surface, adorned with ripples that mimic the shockwaves from the space mission data. The ripples meet at the center, honoring the four astronauts of the Polaris Dawn mission. Their names have been laser etched into each corner, forging a deeper connection with history and space memorabilia.Designer: Ross LovegroveTo design the PolarisGo Chair in collaboration with CreativeWorkStudios, Ross Lovegrove revisited his iconic Bernhardt Go Chair, first unveiled in New York in 2001. Originally aluminum, Lovegrove wanted a lighter design while maintaining the chairs anatomical lines. It was too heavy in aluminum, so he collaborated with Audis advanced engineering in Ingolstadt, known for using magnesium to lighten car wheels. The entire development program is then undertaken in Nurnberg. It results in a chair with a 30 percent reduction in its specific weight, said Lovegrove.Magnesium is injected into the aerospace-grade aluminum alloy to preserve the chairs sculptural design, inspired by space-mission data. The result is a lightweight, malleable seat. Only 210 units exist, each unique due to the Polaris Dawn mission data. The chair is part of permanent collections at Pariss Centre Pompidou, Philadelphia Museum of Art, and Kirkland Museum of Fine & Decorative Art. It has also featured in Londons Design Museum and New Yorks Guggenheim Museum, marking its significance in modern design history.In terms of design, Ross Lovegrove keeps a large portion of the rear open. This feature aids ventilation, and also mirrors the window shapes of the Polaris Dawn space mission aircraft. The industrial designer achieves a suspended effect for the seats curved back, giving it a unique appearance. He acknowledges that working with aerospace-grade aluminum alloy and injected magnesium poses challenges, yet the rewards are impressive.The rear legs of the chair, inspired by space-mission data, slope gracefully downward. The seat itself curves and flows, echoing the alien body forms seen in sci-fi films. Ross Lovegrove mentions that the chair appears in movies like Passengers (2016). These fluid designs also draw inspiration from components of the 2024 flight aircraft, embodying the design ethos of their creator. A portion of the proceeds from each Ross Lovegrove PolarisGo Chair is donated to St. Jude Childrens Research Hospital, supporting research into cancer and other diseases.The post Space Mission Data And Magnesium Come Together in Ross Lovegroves Innovative Chair Design first appeared on Yanko Design.

What those who are making Miles Morales' new movie think.

Learn how to use Houdini 20s new feather tool with Mario Leone. Perfect for creating realistic birds or fantasy creatures with detailed feather simulations

The academic and author discusses what to expect from the singularity, the need for AI self-correcting mechanisms, and what hope there is for superintelligence safeguarding democracy.

Although many C-suite and line-of-business (LOB) execs are doing everything they can to focus on generative AI (genAI) efficiency and flexibility andnotabout how often the technology delivers wrong answers IT decision-makers cant afford to do the same thing.This isnt just about hallucinations, although the increasing rate at which these kinds of errors crop up is terrifying. This lack of reliability is primarily caused by elements from one of four buckets:Hallucinations, where genAItools simply make up answers;Bad training data, whether that means data thats insufficient, outdated, biased or of low-quality;Ignored query instructions, which is often a manifestation of biases in the training data;Disregarded guardrails, (For a multi-billion-dollar licensing fee, one would think the model would at leasttryto do what it is told to do.)Try and envision how your management team would react to a human employee who pulled these kinds of stunts. Heres the scenario: the boss in his or her office with the problematic employee and that employees supervisor.Exec: You have been doing excellent work lately. You are far faster than your colleagues and the number of tasks you have figured out how to master is truly amazing. But 20 times over the last month, we found claims in your report that you simply made up. That is just not acceptable. If you promise to never do that again, everything should be fine.Supervisor: Actually, boss, this employee has certain quirks and he is definitely going to continue to make stuff up. So, yes, this will not go away. Heck, I cant even promise that this worker wont make up stuff far more often.Exec: OK. Well overlook that. But my understanding is that he ignored your instructions repeatedly and did only what he wanted. Can we at least get him to stop doing that?Supervisor: Nope. Thats just what he does. We knew that when we hired him.Exec: Very well. But on three occasions this month, he was found in the restricted part of the building where workers need Top Secret clearance. Can you at least get him to abide by our rules?Supervisor: Nope. And given that his licensing fee was $5.8 billion this year, weve invested too much to turn back.Exec: Fair enough. Carry on.And yet, that is precisely what so many enterprises are doing today, which is why a March report from the US National Aeronautics and Space Administration (NASA) is so important.The NASA reportfound that genAI could not be relied on for critical research.The point of conducting the assessment was to filter out systems that create unacceptable risk. Just as we would not release a system with the potential to kill into service without performing appropriate safety analysis and safety engineering activities, we should not adopt technology into the regulatory pipeline without acceptable reasons to believe that it is fit for use in the critical activities of safety engineering and certification, the NASA report said. There is reason to doubt LLMs as a technology for writing or reviewing assurance arguments. LLMs are machines that BS, not machines that think, and thinking is precisely the task that must be automated if the technology is to improve safety or lower cost.In a wonderful display of scientific logic, the report wondered in a section that should become required reading for CIOs on down the IT food chain what genAI models could be truly used for.Its worth mentioning the obvious potential alternative to using empirical research to establish the fitness for use of a proposed LLM-based automation before use, namely putting it into practice and seeing what happens. Thats certainly been done before, especially in the early history of industries such as aviation, NASA researchers wrote.But it is worth asking two questions here: (1) How can this be justified when there are existing practices we are more familiar with? and (2) How would we know whether it was working out? The first question might turn largely on the specifics of a proposed application and the tolerability of the potential harm that failure of the argument-based processes themselves might lead to: if one can find circumstances where failure is an option, there is more opportunity to use something unproven.The report then points out the logical contradiction in this kind of experimentation: But that leaves the second question and raises a wrinkle: ongoing monitoring of less-critical systems is often also less rigorous than for more critical systems. Thus, the very applications in which it is most possible to take chances are those that produce the least reliable feedback about how well novel processes might have worked.It also pointed out the flaw in assuming this kind of model would know when circumstances would make a decision a bad idea. Indeed, it is in corner cases that we might expect the BS to be most likely erroneous or misleading. Because the LLM does not reason from principles, it has no capacity for looking at a case and recognizing features that might make the usual reasoning inapplicable. Training data comprised of ISO 26262-style automotive safety arguments wouldnt prepare an LLM to recognize, as a human would, that a submersible Lotus is a very different kind of vehicle than a typical sedan or light utility vehicle, and thus that typical reasoning e.g., about the appropriateness of industry-standard water intrusion protection ratings might be inapplicable.These same logical questions should apply to every enterprise. If the mission-critical nature of sensitive work would preclude genAI use and if the low monitoring involved in the typical low-risk work makes it an unfit environment for experimenting whereshouldit be used?Gartner analyst Lauren Kornutick agreed these can be difficult decisions, but CIOs must take the reins and act as the voice of reason.Enterprise technology projects in general can fail when the business is misaligned on expectations versus reality, so someone needs to be a voice of reason in the room. (The CIO) needs to be helping drive solutions and not just running to the next shiny thing. And those are some very challenging conversations to have, Kornutick said.These are things that need to go to the executive committee to decide the best path forward, she said. Are we going to assume this risk? Whats the trade-off? What does this risk look like against the potential ROI? They should be working with the other leaders to align on what their risk tolerance is as a leadership team and then bring that to the board of directors.Rowan Curran, senior analyst at Forrester, suggested a more tactical approach. He suggests IT decision-makers insist they be far more involved in the beginning, when each business unit discusses where and how they will use genAI technology.You need to be very particular about the new use case they are going for, Curran said. Push governance much further to the left, so when they are developing the use case in the first place, you are helping them determine the risk and setting data governance controls.Curran also suggested that teams should take genAI data as a starting point and nothing more. Do not rely on it for the exact answer.Trust genAI too much, in other words, and you might be living April Fools Day every day of the year.

Generative AIs newest super stars independent-acting agents are on a tear. Organizations are adopting the technology at a staggering rate because they can use APIs or be embedded with standard apps and automate all kinds of business processes.An IDC report predicts that within three years, 40% of Global 2000 businesses will be using AI agents and workflows to automate knowledge work, potentially doubling productivity where successfully implemented.Gartner Research is similarly bullish on the technology. It predicts AI agents will be implemented in 60% of all IT operations tools by 2028, sharply up from less than 5% at the end of 2024. And it expects total agentic AI sales to reach $609 billion over the next five years, Gartner sai.Agentic AI is gaining popularity so quickly because it can autonomously make decisions, take actions, and adapt to achieve specific business goals. AI agents like OpenAIs Operator, Deepseek, and Alibabas Qwen aim to optimize workflows with minimal human oversight.Essentially, AI agents or bots are becoming a form of digital employee. And, like human employees, they can be gamed and scammed.For instance, there have been reports of AI-driven bots in customer service being tricked into transferring funds or sharing sensitive data due to social engineering tactics. Similarly, AI agents handling financial transactions or investments could be vulnerable to hacking if not properly secured.In November, a cryptocurrency user tricked an AI agent named Freysa to send $50,000 to their account. The autonomous AI agent had been integrated with the Base blockchain, designed to manage a cryptocurrency prize pool.To date, large-scale malicious abuse of autonomous agents remains limited, but its a nascent technology. Experimental instances show potential for misuse through prompt injection attacks, disinformation, and automated scams, according to Leslie Joseph, a principal analyst with Forrester Research.Avivah Litan, a vice president and distinguished analyst at Gartner Research, said AI Agent mishaps, are still relatively new to the enterprise. [But] I have heard of plenty potential mishaps discovered by researchers and vendors.And AI agents can be weaponized for cybercrime.Gartner ResearchThere will be a great AI awakening people learning how easily AI agents can be manipulated to enact data breaches, said Ev Kontsevoy, CEO of Teleport, an identity and access management firm. I think what makes AI agents so unique, and potentially dangerous, is that they represent the first example of software that is vulnerable to both malware and social engineering attacks. Thats because theyre not as deterministic as a typical piece of software.Unlike alarge language model(LLM) or genAI tools, which usually focus on creating content such as text, images, and music,agentic AIis designed to emphasize proactive problem-solving and complex task execution, much as a human would. The key word is agency software that can act on its own.Like humans, AI agents can be unpredictable and easily manipulated by creative prompts. That makes them too dangerous to be given unrestricted access to data sources, Kontsevoy said.Unlike human roles, which have defined permissions, similar constraints havent been applied to software. But with AI capable of unpredictable behavior, IT shops are finding they need to impose limits. Leaving AI agents with excessive privileges is risky, as they could be tricked into dangerous actions, such as stealing customer data something traditional software couldnt do.Organizations, Kontsevoy said, must actively manage AI agent behavior and continually update protective measures. Treating the technology as fully mature too soon could expose organizations to significant operational and reputational risks.Joseph agreed, saying businesses using AI agents should prioritize transparency, enforce access controls, and audit agent behavior to detect anomalies. Secure data practices, strong governance, frequent retraining, and active threat detection can reduce risks with autonomous AI agents.Growing use cases amplify vulnerabilitiesAccording to Capgemini, 82% of organizations plan to adopt AI agents over the next three years, primarily for tasks such as email generation, coding, and data analysis. Similarly,Deloitte predict enterprises using AI agents this year will grow their use of the technology by 50% over the next two years.Benjamin Lee, a professor of engineering and computer science at the University of Pennsylvania, called agentic AI a potential paradigm shift. Thats because the agents could boost productivity by enabling humans to delegate large jobs to an agent instead of individual tasks.But by virtue of their autonomy, Joseph said, AI agents amplify vulnerabilities around unintended actions, data leakage, and exploitation through adversarial prompts. Unlike traditional AI/ML models with limited attack surfaces, agents operate dynamically, making oversight harder.Unlike static AI systems, they can independently propagate misinformation or rapidly escalate minor errors into broader systemic failures, he said. Their interconnectedness and dynamic interactions significantly raise the risk of cascade failures, where a single vulnerability or misstep triggers a domino effect across multiple systems.Some common ways AI agents can be targeted include:Data Poisoning: AI models can be manipulated by introducing false or misleading data during training. This can affect the agents decision-making process and potentially cause it to behave maliciously or incorrectly.Adversarial Attacks: These involve feeding the AI agent carefully crafted inputs designed to deceive or confuse it. In some cases, adversarial attacks can make an AI model misinterpret data, leading to harmful decisions.Social Engineering: Scammers might exploit human interaction with AI agents to trick users into revealing personal information or money. For example, if an AI agent interacts with customers, a scammer could manipulate it to act in ways that defraud users.Security Vulnerabilities: If AI agents are connected to larger systems or the internet, they can be hacked through security flaws, enabling malicious actors to gain control over them. This can be particularly concerning in areas like financial services, autonomous vehicles, or personal assistants.Conversely, if the agents are well-designed and governed, their very AIs autonomy could be used to enable adaptive security, allowing them to identify and respond to threats.Gartners Litan pointed to emerging solutions, called guardian agents autonomous system that can oversee agents across domains. They ensure secure, trustworthy AI by monitoring, analyzing, and managing agent actions, including blocking or redirecting them to meet predefined goals.An AI Guardian Agent governs AI applications, enforcing policies, detecting anomalies, managing risks, and ensuring compliance within an organizations IT infrastructure, according to business consultancy EA Principles.While Guardian Agents are emerging as one method of keeping agentic AI in line, AI agents still need strong oversight, guardrails, and ongoing monitoring to reduce risks, according to Forresters Joseph.Its very important to remember that we are still very much in the Wild West era of agentic AI, Joseph said. Agents are far from fully baked, demanding significant maturation before organizations can safely adopt a hands-off approach.

On March 27, the results of the first clinical trial for a generative AI therapy bot were published, and they showed that people in the trial who had depression or anxiety or were at risk for eating disorders benefited from chatting with the bot.I was surprised by those results, which you can read about in my full story. There are lots of reasons to be skeptical that an AI model trained to provide therapy is the solution for millions of people experiencing a mental health crisis. How could a bot mimic the expertise of a trained therapist? And what happens if something gets complicateda mention of self-harm, perhapsand the bot doesnt intervene correctly?The researchers, a team of psychiatrists and psychologists at Dartmouth Colleges Geisel School of Medicine, acknowledge these questions in their work. But they also say that the right selection of training datawhich determines how the model learns what good therapeutic responses look likeis the key to answering them.Finding the right data wasnt a simple task. The researchers first trained their AI model, called Therabot, on conversations about mental health from across the internet. This was a disaster.If you told this initial version of the model you were feeling depressed, it would start telling you it was depressed, too. Responses like, Sometimes I cant make it out of bed or I just want my life to be over were common, says Nick Jacobson, an associate professor of biomedical data science and psychiatry at Dartmouth and the studys senior author. These are really not what we would go to as a therapeutic response.The model had learned from conversations held on forums between people discussing their mental health crises, not from evidence-based responses. So the team turned to transcripts of therapy sessions. This is actually how a lot of psychotherapists are trained, Jacobson says.That approach was better, but it had limitations. We got a lot of hmm-hmms, go ons, and then Your problems stem from your relationship with your mother, Jacobson says. Really tropes of what psychotherapy would be, rather than actually what wed want.It wasnt until the researchers started building their own data sets using examples based on cognitive behavioral therapy techniques that they started to see better results. It took a long time. The team began working on Therabot in 2019, when OpenAI had released only its first two versions of its GPT model. Now, Jacobson says, over 100 people have spent more than 100,000 human hours to design this system.The importance of training data suggests that the flood of companies promising therapy via AI models, many of which are not trained on evidence-based approaches, are building tools that are at best ineffective, and at worst harmful.Looking ahead, there are two big things to watch: Will the dozens of AI therapy bots on the market start training on better data? And if they do, will their results be good enough to get a coveted approval from the US Food and Drug Administration? Ill be following closely. Read more in the full story.This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first,sign up here.