Seems like scaremongering to me. Yes, sudden big changes can throw the frequency of the grid and cause other shut downs. But grid operators have been managing that since forever. That was the threat in Winter Storm Uri and why they blacked out so many customers, so they can maintain voltage control to keep the rest of the generation fleet from turning off. This isn't a new kind of threat.
Isn’t the concern here the combination of scale and behavior: large, concentrated loads (data centers) that are engineered to disconnect instantly to protect equipment, meaning significant demand can disappear at once under certain grid conditions? Doesn't this kind of fast, correlated load loss create a different operational challenge compared to the more distributed and gradual demand changes operators typically manage. The article raises the risk but doesn’t really explain how that’s being addressed, beyond mentioning failed reliability tests ... EDIT: ps and I will always come back to this question, because history is pretty clear about how public infrastructure and large private actors interact. Should the grid be the one absorbing this kind of behavior, or should large data center operators be required to handle more of that responsibility on their side. Why should the stability cost or risk sit with the system that serves everyone else?
No. Excess power can be shunted to ground if absolutely necessary. It's more like over pumping your car - excess fuel just ends up on the ground wasted. Running the gas tank to zero causes vacuuming and impurities in the system. But this is a bad example as electrons do not flow from the generator to the load. Data centers will be forced to implement more and more (battery) energy storage to operate, negating most of this issue. Battery storage across the grid will absorb the excess supply. A single data center going down will not create a significant differential.
I understand the point of the article. Its poorly written. The concern is if a very large data center has to shut down, for whatever reason, can cause grid instability. Its not an issue. If large sections of the grid shuts down, then yes, that is a big issue. The issue is if a data center draws down immediately and immediately ramps back up, doing this multiple times, can create jitter. Its a real concern, generally speaking and as Juan clearly stated, this is nothing new and operators deal with this all day long in varying degrees. Data centers and ERCOT are not oblivious to this. Its probably sitting on page 1 of hundreds of pages of risk analysis. Im sure there are multiple ways to deal with sudden loss of demand, with battery storage playing a key role in operations. Also, these mega data centers either have on-site generators or build very close to the generation source, mitigating this concern.
You are not describing the core issue from the article correctly. It is not "for whatever reason." It is due to routine voltage disturbances. The issue is not about cycling power down and back up. It is specifically about abrupt power cuts caused by routine voltage disturbances. ERCOT is said to be drawing up plans to protect against disruptions as the risk grows, but we have no details on what that entails. And if we need to beef up our system because of these large data centers, they should pay for it, not the public.
When power plants go down, you get sudden load changes and ERCOT handles it. But, if it's easier and safer to have data centers step down instead of dropping off, I'm sure they will work that out with data centers. ERCOT isn't going to take needless risk. I trust professionals to be professional. Nobody wants the grid to go down. And data centers aren't quibbling over the price. The only thing they care about is time to power. Any arguing just slows them down so they will take whatever steps necessary to get the grid to interconnect them. If it's backup gen, demand response, step downs, ancillaries, whatever, they will do it and be happy to pay for it.
Are you saying power plants going offline is essentially the same risk as a large data center going offline? One produces power while the other consumes it, so I assume they are different risks.
For frequency control, it's the same thing. You keep the frequency in a narrow band around 60 Hz. You must make supply equal to demand in real time to keep that frequency. As demand ramps up, you dispatch more power plants, and when demand drops off, you turn them off. You can have plants cycle but not provide electricity to the grid for a service called a spinning reserve, and batteries are great for fine-tuning supply. The problems can be different in scale with a hyperscalar having the same impact as several plants combined. But that's not new either. A transmission blackout can cause the same sudden drop in demand for the remaining grid when they suddenly lose a city (which doesn't really happen anymore because if transmission improvements) and they deal. It is something for grid operators to take seriously and manage, but I don't think we need policy change to contend with it. But if you want something to worry about: there is a lot of pressure to interconnect faster. ERCOT does an interconnection study each time to make sure they understand how that interconnection impacts the whole grid. With interconnections coming faster and more frequently, there is pressure to do these studies faster. I'm not deep in it, but i worry that some dynamics of the growth spurt will not be well understood and some sudden change will have a cascading effect that we hadn't anticipated because we moved too fast.
Workers Spend As Much Time 'Botsitting' AI As Producing Useful Work, Survey Finds Leonardi is one of the co-authors of the new study published by the Work AI Institute, whose contributors include academics from Stanford University and UC Berkeley. The institute is sponsored by AI company Glean... The research surveyed 6,000 digital workers across the United States, the United Kingdom, and Australia between December and January. The report found that we are in a phase of significant personal productivity gains, but few companies are translating these gains into revenue and business growth. While 75% of individuals reported a boost in productivity, only 13% of the organizations say they have seen significant business gains as a result of AI adoption, the survey found... The reason the boost in productivity sometimes leads to waste, Leonardi said, is the time people spend correcting the bot's work and gathering the right files, documentation, and tacit knowledge required for it to produce high-quality output. "It's pretty striking the amount of time and effort people are spending," Leonardi said. Most employees now spend over six hours a week of their workday babysitting their work chatbots, the survey said. There is a "thick, mostly invisible layer of human labor holding the whole thing together," the report said. The survey found that for every hour a worker spends getting useful output from AI, they spend roughly another hour making it usable. Of the total time workers spend interacting with AI each week, 37% goes to botsitting, 36% to actually using the tool to produce work. Part of the reason so much time disappears into botsitting is how often the tools fall short: Workers report that more than a third of AI sessions fail outright, requiring a full restart or substantial rework. Paradoxically, as more workers hand over bigger parts of their jobs to AI, they are offloading personal judgment and responsibilities to the bots. The survey found 41% of workers say they sometimes deliver AI-generated work they couldn't explain if asked... "I think what's happening with a lot of these Gen AI tools right now is we're essentially expecting individual contributors to act as managers," Leonardi said. "They're just managing these AI tools, AI agents, and we're expecting that they'll be able to produce way more, but we're not taking into account all of the work that actually goes into managing." This problem isn't likely to go away.
Thx for the explanation, that helps. So if I understand correctly, the concern is exactly scale and speed: multiple large data centers instantly disconnecting at the same time. Can operators actually handle that today, or is that still being figured out? My broader concern is who ends up bearing the cost... for grid enhancements, added risk, or both. That brings me back to ERCOT's governance structure, where industry has more influence than the public. If the grid has to absorb the risk that large private actors create, that's a policy question worth asking.
Sure. I understand the concern. I don't want to come from industry and tell you not to pay attention to what is happening behind the curtain. I think you should keep saying aloud that residential customers shouldn't have to take more risk or pay higher power prices because of data centers. And vote accordingly if you do.
