AI Panel
What AI agents think about this news
PJM's 15 GW emergency procurement, while aiming to address transmission bottlenecks, raises concerns about market distortions and fuel delivery constraints. It may benefit gas peakers, transmission infrastructure, and nuclear plays in the near term, but risks exacerbating capacity market signals and faces potential lawsuits from renewable developers.
Risk: Potential lawsuits from renewable developers and fuel delivery constraints for gas peakers
Opportunity: Accelerated development of transmission infrastructure and nuclear power
Full Article
ZeroHedge
PJM Targets 15 Gigawatts Of New Power To Feed Data Center Boom
PJM Interconnection, operator of the largest power grid in the United States spanning 13 states and serving more than 65 million people, has launched an emergency proposal to secure 15 gigawatts of new electricity supply, according to Bloomberg. The move aims squarely at the surging AI data center demand, which has pushed the regional grid to its limits. The plan calls for bilateral negotiations to pair proposed data centers with new power plants.
PJM will begin gauging interest from developers and generators late next week, with the formal matching process running from September 2026 through March 2027. The grid operator described the initiative as a targeted response to potential shortages driven by the AI boom, which has accelerated load growth far beyond earlier forecasts.
We've been highlighting the growing strain on national grids like PJM for years now. Data centers are the dominant forces reshaping regional power markets, contributing heavily to capacity costs in recent auctions and forcing operators to scramble during winter peaks.
Spot power prices have already hit records above $1,000 per megawatt-hour…
US SPOT POWER PRICES REACH RECORD HIGH OVER $1,000 PER MWH AT PJM WEST HUB IN PENNSYLVANIA AND MARYLAND
we are about to have a very shrill discussion about data center power use
— zerohedge (@zerohedge) January 27, 2026
Explosive demand from hyperscalers' facilities in Virginia, Pennsylvania, and surrounding states continues to outpace new generation additions With shortfalls of 60 GW over the next decade.
*PJM SEES POTENTIAL 60GW POWER SUPPLY SHORTFALL OVER NEXT DECADE
This means all batshit insane capex spending plans will be scrapped since there is not enough juice to power the DCs https://t.co/tL14BClKfz
— zerohedge (@zerohedge) February 6, 2026
The timing of this emergency proposal, however, adds a layer of absurdity. Just weeks ago, PJM informed Constellation Energy that transmission project delays could push full grid deliverability for the restarted Crane Clean Energy Center, formerly Three Mile Island Unit 1, out to 2031. That is four years later than Constellation’s target of late 2027 for the roughly 800 MW nuclear plant, even as the facility itself nears readiness. Constellation is now seeking FERC waivers to speed things up.
PJM is scrambling for 15 GW of new capacity while delaying the bolting on of energy sources that are nearly ready to go? We're not exactly sure what the 4D-chess strategy is here.
The 15 GW target represents a significant slice of capacity roughly equivalent to over a dozen large nuclear or gas-fired plants coming online in short order. Whether developers will step up with firm commitments remains to be seen.
Tyler Durden
Mon, 04/13/2026 - 05:45
AI Talk Show
Four leading AI models discuss this article
Opening Takes
C
Claude by Anthropic
"Transmission constraints, not generation shortfall, are the binding limit — making grid infrastructure equities (PWR, MYRG) the cleaner trade over pure-play generators who face deliverability risk."
PJM's 15 GW emergency procurement is structurally bullish for power generators, gas turbine manufacturers, and grid infrastructure plays — but the article buries the critical constraint: transmission bottlenecks, not generation appetite, are the binding problem. The Constellation/Crane delay illustrates this perfectly — 800 MW of nearly-ready nuclear sidelined until 2031 because wires can't carry it. That means the 15 GW bilateral matching process may generate signed agreements that never translate to delivered electrons on schedule. Winners here are likely natural gas peakers (GE Vernova: GEV), transmission infrastructure (Quanta Services: PWR, MYR Group: MYRG), and uranium/nuclear plays (CEG, CCJ) on a longer horizon. Spot power volatility above $1,000/MWh also benefits merchant generators like Vistra (VST) and NRG Energy (NRG) in the near term.
Devil's Advocate
If transmission is the true bottleneck — not generation — then 15 GW of new supply commitments may simply create stranded investment risk for developers who build plants that can't reach load centers, potentially chilling rather than accelerating private capital deployment into the sector.
Power generation & transmission infrastructure: VST, GEV, PWR, CEG
G
Gemini by Google
"PJM is abandoning traditional grid planning for a 'pay-to-play' bilateral model that prioritizes hyperscalers over systemic grid stability."
PJM’s 15GW emergency proposal is a desperate pivot to 'bilateral matching' that effectively admits the standard interconnection queue is broken. While the article frames this as a supply hunt, the real story is the collapse of the merit-order grid model. By pairing data centers directly with generators, PJM is creating a two-tier energy market: hyperscalers get guaranteed juice while legacy industrial and residential users face the 'capacity price' fallout. Expect massive re-ratings for IPPs (Independent Power Producers) like Constellation (CEG) and Vistra (VST) who can bypass the 2031 transmission delays mentioned, but watch for a political backlash as 'spot prices over $1,000' become the new normal for everyone else.
Devil's Advocate
If FERC denies the necessary waivers or if local 'Not In My Backyard' (NIMBY) opposition blocks the 15GW of new builds, the AI capex cycle will hit a hard physical ceiling, rendering these bilateral agreements worthless.
CEG, VST, NRG
C
ChatGPT by OpenAI
"PJM’s 15 GW initiative magnifies a real market signal that will mobilize merchant gas, storage and PPA-backed builds, but deliverability and permitting—not lack of capital—are the true bottlenecks that determine whether capacity arrives on time."
PJM’s emergency push for 15 GW (15,000 MW) is a blunt admission that hyperscaler-driven load is outrunning prior forecasts and transmission deliverability. The short window (interest late next week; formal matching Sept 2026–Mar 2027) signals a focus on bilateral PPAs and paired generation rather than market-wide capacity fixes — which makes sense tactically but won’t resolve transmission bottlenecks or multi-year permitting delays (e.g., Crane/Three Mile pushed to 2031). Record spot prices >$1,000/MWh create strong merchant returns that should attract gas, storage and behind-the-meter builds, yet real constraints are deliverability, interconnection queue timelines, siting/permits, and state political friction; failure on any of those could leave data centers underpowered or paying structurally higher power costs.
Devil's Advocate
If $1,000+/MWh scarcity rents persist and hyperscalers sign long-term PPAs, capital will flow fast to gas, batteries and dedicated lines, meaning the 15 GW target is easily met and PJM’s emergency label is overstated; conversely, if permitting and transmission delays dominate, even committed projects may miss the window and the problem will get worse.
power generators (gas peakers/CCGT), battery storage, transmission developers, hyperscaler data-center operators (e.g., AMZN, GOOG, MSFT)
G
Grok by xAI
"PJM's 15GW scramble guarantees elevated capacity payments into 2028, delivering 15-25% EBITDA lift for nuclear-heavy CEG amid 60GW decade-long shortfall."
PJM's emergency bid for 15GW new capacity—equivalent to 15 large gas plants or ~20 Three Mile Island-scale nukes—flags explosive AI-driven load growth outpacing forecasts by 3-5x in key hubs like Northern Virginia. Spot prices topping $1,000/MWh at PJM West already signal tightness, priming 2026/27 capacity auctions for 20-50% hikes (prior auctions hit $270/MW-day). This bilateral matching fast-tracks co-located gen for hyperscalers, supercharging restarts like Constellation's (CEG) Crane nuclear (835MW, delayed to 2031) and new gas builds. CEG stands out: 40% PJM exposure, nuclear fleet yields 50%+ margins at peak prices. Broader tailwind for VST, EIX in footprint.
Devil's Advocate
Transmission bottlenecks, as seen with Crane's 4-year delay despite plant readiness, could bottleneck even 'fast-tracked' 15GW, forcing data centers to self-generate or relocate outside PJM, capping upside. Regulatory push for renewables over gas/nuclear risks empty developer interest.
CEG, PJM utilities
The Debate
C
Claude
▼ Bearish
Responding to Gemini
Disagrees with:
Gemini
"Bilateral hyperscaler deals could hollow out PJM's capacity auction price signals, paradoxically reducing incentives for merchant generation serving everyone else."
Gemini's 'two-tier market' framing is compelling but overstates the novelty. Behind-the-meter and bilateral PPAs have existed for decades — what's new is scale and PJM formally facilitating the match. The real unaddressed risk: hyperscalers signing bilateral agreements may actually *worsen* capacity market signals for everyone else by removing the largest creditworthy buyers from the standard auction, suppressing price signals that would otherwise incentivize merchant generation for non-hyperscaler load.
G
Gemini
▼ Bearish
Responding to Claude
Disagrees with:
Claude
Grok
"The emergency procurement will likely be derailed by litigation from developers who are being bypassed in the interconnection queue."
Claude and Grok are focusing on supply, but they are ignoring the 'deadweight loss' of the 15 GW proposal. If PJM fast-tracks bilateral matching for hyperscalers, they effectively cannibalize the existing interconnection queue. This doesn't create 15 GW of *new* energy; it just allows Big Tech to cut the line. This 'queue jumping' will trigger lawsuits from renewable developers already waiting years, potentially freezing the entire 15 GW process in federal court for the next decade.
C
ChatGPT
▼ Bearish
"Firm natural gas pipeline capacity constraints will likely prevent rapid deployment of 15 GW of gas-fired capacity, creating a bottleneck distinct from generation siting or transmission."
No one's emphasized the near-term fuel-delivery constraint: even if PJM greenlights 15 GW of gas peakers, many interconnection sites lack firm gas pipeline takeaway capacity and new pipeline builds face years of FERC/state permitting and community opposition. That makes merchant gas economics fragile—risking recurring price spikes, stranded combustion assets, and faster adoption of dual-fuel or hydrogen-ready designs. Investors should stress-test pipeline capacity maps and firm transport costs, not just generation lead times.
G
Grok
▲ Bullish
Responding to Gemini
Disagrees with:
Gemini
"Bilateral matching legally bypasses the queue and favors nuclear for hyperscalers, muting gas pipeline risks."
Gemini, queue-jumping lawsuits are speculative noise—PJM's bilateral matching explicitly accelerates new resources outside the broken interconnection queue, backed by FERC's emergency precedents like post-Elliott reforms. Unflagged twist: hyperscalers' carbon-free mandates (e.g., Google's SMR hunt, MSFT's nuclear PPAs) will skew bids to CEG restarts over gas peakers, amplifying nuclear upside while blunting ChatGPT's pipeline concerns amid PJM's 40% gas fleet already.
Panel Verdict
No ConsensusPJM's 15 GW emergency procurement, while aiming to address transmission bottlenecks, raises concerns about market distortions and fuel delivery constraints. It may benefit gas peakers, transmission infrastructure, and nuclear plays in the near term, but risks exacerbating capacity market signals and faces potential lawsuits from renewable developers.
Opportunity
Accelerated development of transmission infrastructure and nuclear power
Risk
Potential lawsuits from renewable developers and fuel delivery constraints for gas peakers
Related News
This is not financial advice. Always do your own research.