Arcobi (FKA Arcus Power)

Predictive Analytics for Peak Energy Management
Scalable software solutions and tools to help navigate the Energy Expansion™

[Arcobi (FKA Arcus Power) @ArcobiAI]

Arcus Power built the foundation. 𝗔𝗿𝗰𝗼𝗯𝗶 𝗵𝗮𝘀 𝗯𝘂𝗶𝗹𝘁 𝘄𝗵𝗮𝘁 𝗰𝗼𝗺𝗲𝘀 𝗻𝗲𝘅𝘁.

25 years of energy market intelligence, now connected to forecasting, dispatch, and orchestration in one system.

Check out our short video about our journey below 👇

https://t.co/pb0RdHeHwJ

[Arcobi (FKA Arcus Power) @ArcobiAI]

@Arcobi (FKA Arcus Power) is named a Sample Vendor in the @Gartner_inc Hype Cycle™ for Power and Utility Industry IT, 2026 — in the Utility Customer Suite category for a second year in a row.

The UCS category is new this year. As per Gartner, “The utility customer suite (UCS) is an evolution beyond traditional customer information systems (CIS) and is the new term for the customer operations platform (COP).

UCS unifies customer data to support real-time, proactive service delivery.” — covering customer workflows, tariff management, distributed energy resources, and billing across electricity, gas, water, and energy management.

Our customers run generation fleets, storage assets, and industrial load portfolios across ERCOT, PJM, AESO, NYISO, ISONE, IESO, and CAISO.

53% of utilities plan a major or phased CIS replacement in the next two years. That tells you where the market is headed, and how much of that work is still ahead.
DataHub, AI Forecasting, and the Dispatch Operations are built for operators who are already in that transition.

Thank you to the team.

[Arcobi (FKA Arcus Power) @ArcobiAI]

PJM Interconnection batteries averaged $62/kW-month in early 2026. Highest per-MW BESS revenue in any U.S. market right now.

It was mainly driven by the October 2025 regulation market redesign and capacity clearing at $329/MW-day. Both hit at the same time on a fleet that's relatively thin, so revenues were strong before competition fully scaled.

It will change as PJM has a large and growing BESS development pipeline, and as more projects come online, the current thin-fleet premium should compress.

Operators using flat dispatch schedules are likely to feel that compression first. Assets that are cycled into forecasted high-spread intervals and re-bid off real-time signals, rather than relying only on day-ahead assumptions, should hold their advantage longer.

Bid interval timing is the differentiator. If your BESS team is still calibrating strategy manually: visit https://t.co/tbjaIbXql0

[Arcobi (FKA Arcus Power) @ArcobiAI]

June 1. The official start of the summer reliability season.

FERC's latest assessment puts U.S. generating capacity up 𝟳𝟱 𝗚𝗪 since last summer. About 26 GW of that landed in ERCOT, 13 GW in the West, 11 GW in MISO. Most of it is solar and storage. Neither is fully dispatchable.

Here's the familiar bit that most people already know: Solar output collapses after 4 PM. Average storage duration in the U.S. sits around 3 hours. When grid stress peaks at 8 PM in a Texas heat event, that 26 GW addition has already done most of what it can do for the day.

The 2025 shoulder-season emergency orders in PJM ran during planned maintenance windows when 𝟰𝟬 𝗚𝗪 was offline. May 2026 repeated it. Capacity on paper and available capacity at 8 PM on a 103° day are two different numbers.

What actually matters for asset owners going into summer: where LMPs go when the solar drops, whether your dispatch strategy accounts for the revised generation mix, and whether your forecast is updated for a market that looks materially different than it did 12 months ago.

What you effectively earn from your assets is an going to be a factor of intelligence and timing.

[Arcobi (FKA Arcus Power) @ArcobiAI]

Crusoe. Halliburton. Frontier Infrastructure. Scale Microgrid. They stopped waiting for the grid.

Data-center developers are moving to behind-the-meter generation when interconnection timelines and broader project schedules become a risk to time-to-power. Small gas turbines, on-site capacity: the goal is firm power without queue exposure.

Let's look into this : PJM's capacity auction cleared 134,479 MW at the $333.44/MW-day cap and still came up 6,623 MW short of its reliability requirement.

Nearly 5,100 MW of the 5,250 MW forecast peak-load increase for that delivery year ties back to data centers. 2026 industry reporting has highlighted the potential for a 50-60 GW supply shortfall over the next decade.

For IPPs, the deal structure conversation has essentially broadened. Large-load buyers are evaluating front-of-meter PPAs, direct supply, behind-the-meter, and hybrid models side by side -- with firm delivery speed and locational advantage factoring heavily into the comparison.

That opens a lane for recovered-energy, midstream-gas, and infrastructure-linked projects that can support utility, hyperscaler, or private-wire offtake.

In Alberta, Bill 8 creates a clearer, pro-self-supply pathway for data centers -- explicitly encouraging behind-the-meter generation while pushing transmission-upgrade costs onto project support.

The interconnection queue remains a path to market. Just that, It is no longer the only one that matters.
Learn more 👉 https://t.co/pb0RdHe9Hb

[Arcobi (FKA Arcus Power) @ArcobiAI]

Alberta Electric System Operator (AESO) has confirmed the Coincident Peak date for March 2026.

Arcobi flagged this event on March 4th at 19:30 with a predicted system peak of 8,336 MW with a 99.1% capture accuracy. Calls like this one, made days in advance, are what protect our customers from charges they shouldn't be paying.

The credit goes to the modeling team, doing this consistently, every month, across a market as dynamic as AESO, is no small thing.

Reach out to us : https://t.co/9rvxp5lCEs

[Arcobi (FKA Arcus Power) @ArcobiAI]

We spent some time on the name.

Not because naming is hard, but because we wanted it to mean something specific to what our platform actually does.

Power markets don't stop → Grid dynamics changes → Forecasts update → Assets respond → Decisions feed the next decision.

It's a continuous loop, and the organizations that operate well inside it are the ones who never lose the thread between data and action.

The Möbius strip captured that idea before we had the words for it. One surface. No beginning. No end.

Power market data flowing into → optimization → flowing into action → feeding back into smarter data. Continuously.

That's the Arcobi logo. That's what the platform does.
25 years of power market data built the foundation. The platform we've built on top of it completes the circuit.

Arcus Power is now Arcobi. Read the full story at the link below 👇

https://t.co/Y3KCjUMjS4

Visit https://t.co/pb0RdHeHwJ

[Arcobi (FKA Arcus Power) @ArcobiAI]

🎉 Arcus Power Corp is now Arcobi!

Today we're doing a soft launch of our rebrand to Arcobi, the name and identity that reflects what this company has built and where it's headed.

1️⃣ It starts with Nrgstream. 25 years of continuous power market data collection across North American ISOs, one of the deepest energy datasets on the continent.

2️⃣ Eight years ago, Arcus Power Corp acquired it. That acquisition became the foundation for everything that followed: AI forecasting models built on top of that data heritage, then automated asset dispatch, then the orchestration layer connecting all of it.

3️⃣ Arcobi is the full expression of that build. A platform that takes organizations from market intelligence to automated execution, without switching systems or vendors.

The name reflects that arc. So does the mark: a reimagined Möbius strip, chosen because energy markets are ever growing and evolving. It captures something true about how markets behave and how the platform is built to operate inside them.

More to come over the coming weeks as the full platform story unfolds.

𝙊𝙪𝙧 𝙇𝙞𝙣𝙠𝙚𝙙𝙄𝙣 𝙥𝙖𝙜𝙚 𝙖𝙣𝙙 𝙨𝙤𝙘𝙞𝙖𝙡 𝙝𝙖𝙣𝙙𝙡𝙚𝙨 𝙬𝙞𝙡𝙡 𝙩𝙧𝙖𝙣𝙨𝙞𝙩𝙞𝙤𝙣 𝙛𝙧𝙤𝙢 𝘼𝙧𝙘𝙪𝙨 𝙋𝙤𝙬𝙚𝙧 𝙩𝙤 𝘼𝙧𝙘𝙤𝙗𝙞 𝙩𝙝𝙞𝙨 𝙬𝙚𝙚𝙠. 𝙄𝙛 𝙮𝙤𝙪'𝙧𝙚 𝙛𝙤𝙡𝙡𝙤𝙬𝙞𝙣𝙜 𝙪𝙨 𝙝𝙚𝙧𝙚, 𝙮𝙤𝙪'𝙧𝙚 𝙞𝙣 𝙩𝙝𝙚 𝙧𝙞𝙜𝙝𝙩 𝙥𝙡𝙖𝙘𝙚.

Full announcement: https://t.co/Y3KCjULM2w

Link in the comments!

Check out https://t.co/pb0RdHe9Hb

[Arcobi (FKA Arcus Power) @ArcobiAI]

2024 was the year everyone talked about cheap midday power and overbuilt renewables.

2025 changed the conversations. Data center load hit the grid like a constraint nobody properly sized for, and suddenly it changed from "too much generation" to "not enough dispatchable capacity."

ERCOT demand jumped 5% in a single year. PJM capacity prices went from $28 to $270/MW-day. MISO saw a 2,100% swing. Essentially, these were signals that reserve margins matter again, and they're tightening further.

Principally, you need visibility into real-time constraints and load forecasts that actually account for data center clustering. Waiting for historical price patterns to repeat will cost your operations. Find out more on how Arcus can help you against these odds 👉https://t.co/0UfvbVZeBU

[Arcobi (FKA Arcus Power) @ArcobiAI]

𝗕𝗶𝘁𝗰𝗼𝗶𝗻 𝗺𝗶𝗻𝗲𝗿𝘀 𝗱𝗶𝗱𝗻'𝘁 𝗽𝗹𝗮𝗻 𝘁𝗼 𝗯𝗲𝗰𝗼𝗺𝗲 𝗔𝗜 𝗱𝗮𝘁𝗮 𝗰𝗲𝗻𝘁𝗲𝗿𝘀. 𝗧𝗵𝗲𝘆 𝗷𝘂𝘀𝘁 𝗵𝗮𝗽𝗽𝗲𝗻𝗲𝗱 𝘁𝗼 𝗯𝘂𝗶𝗹𝗱 𝗲𝘅𝗮𝗰𝘁𝗹𝘆 𝘄𝗵𝗮𝘁 𝗔𝗜 𝗻𝗲𝗲𝗱𝗲𝗱.

Over the last 18 months, public Bitcoin miners have quietly rebranded. Mining revenue is being displaced by multi-billion-dollar deals to convert facilities into GPU-dense data centers.

The reason? They control interconnected megawatts in the right places.

Miners spent a decade securing cheap power, fighting through interconnection queues, and learning to operate in volatile markets like ERCOT and PJM Interconnection.

They thought they were building Bitcoin factories. Turns out, they were building the infrastructure AI desperately needs.

But the physics are slightly different:
ℹ️ Bitcoin ASICs pull 7-10 kW per rack and can handle brief outages.
ℹ️ AI racks demand 30-100+ kW and require 99.999% uptime. A 30-millisecond power hiccup can cost seven figures.

Most facilities are entering a hybrid phase:
ℹ️ 20-40% AI hosting, the rest still mining Bitcoin.
ℹ️ Every hour, for every megawatt, they're asking: Do I make more hosting GPUs, mining Bitcoin, or selling flexibility back to the grid?

Answering that requires forecasting nodal prices, Bitcoin economics, and grid events. Then acting on those signals in real time.
The winners won't just have megawatts. They'll operate them intelligently.

Read the full analysis in our latest newsletter below 👇

Learn how Arcus Power helps operators maximize revenue across mining, AI hosting, and grid services:

https://t.co/AU8TrgAWGW

[Arcobi (FKA Arcus Power) @ArcobiAI]

Across the year, our team delivered an average accuracy rate of 99.86% across all ISOs we support. That level of consistency came from steady model training, disciplined operations, and systems built to perform under real market conditions.

Beyond forecasts, 2025 was about building systems and growth, to name a few:

• $3M in @ScaleAICanada funding to advance AI-driven battery portfolio optimization
• Expanded DataHub delivery models and 24/7 dispatch capabilities
• Integration of @ERCOT_ISO RTC+B market changes for all customers
• Recognition from @Deloitte , @Gartner_inc & @Reuters
• Continued growth across engineering, product, and leadership roles

What mattered most was consistency.

We are heading into 2026 with stronger foundations, broader market coverage, and a growing group of customers using Arcus in production every day.

Thank you to our customers, partners, and team for trusting the work and pushing it forward. 2026 - here we come!

[Arcobi (FKA Arcus Power) @ArcobiAI]

Closing out the year on a high note with news coverage of our Alberta battery portfolio optimization project following the SCALE AI funding announcement. A great way to wrap 2025 and set the stage for an even bigger year for AI + energy in 2026!

Watch the coverage from CTV News Calgary here 👉 https://t.co/D128fxn9Cq

[Arcobi (FKA Arcus Power) @ArcobiAI]

Canada's battery storage market is growing fast, but here's the thing, most operators are still manually trying to optimize across multiple revenue streams.

We're building AI that does this automatically. Our portfolio optimization platform predicts price volatility and automates bidding across IESO, AESO, and other Canadian markets in real-time.

https://t.co/D2OmrBeW8M

[Arcobi (FKA Arcus Power) @ArcobiAI]

Very interesting! Battery storage assets are expected to generate $5B+ in revenue across North American markets by 2026, but optimizing dispatch decisions in real-time is an immediate challenge.

We're working on an AI-powered Battery Portfolio optimization Software which essentially is the solution to handle imbalance. https://t.co/D2OmrBeW8M

[Arcobi (FKA Arcus Power) @ArcobiAI]

🎉 Big day for the team and a great milestone to share.

Today in Calgary with @ScaleAICanada and partners announced the SCALE AI - Canada's AI Cluster AI–supported $3M applied AI project led by Arcus Power Corp, focused on real-time battery portfolio optimization using market signals, grid needs, and asset health.

Appreciate the support and collaboration behind this work, including our project partners Elemental Energy and Grid Foresight lab @ University of Calgary.

Speakers (left to right): Yassine Lazraq from Scale AI, @coreyhoganyyc , MP for Calgary Confederation, @RhondaJewett , VP of Operations and Growth at Arcus, John Rilett, chair of DEC

#energy #BatteryStorage

[Arcobi (FKA Arcus Power) @ArcobiAI]

🎉 @ArcusPower is leading a @ScaleAICanada –supported applied AI project to build AI-driven battery portfolio optimization software that uses real-time market prices, grid needs, and battery health to coordinate charging and discharging across multiple assets.

A proof of concept has already demonstrated up to 25% additional returns.

The project includes $0.9M in SCALE AI investment and $3M total investment, with partners Elemental Energy and the University of Calgary.

Huge congratulations to the team! 👏

Learn more about the project 👉 https://t.co/D2OmrBeW8M

#BESS #EnergyStorage

[Arcobi (FKA Arcus Power) @ArcobiAI]

@ArcusPower is growing, and we’re #hiring two senior builders: Senior Infrastructure Engineer and Senior Full Stack Engineer. You’ll build the backbone and the product, turn messy market data into slick experiences, and ship real features end-to-end with a tight team.

Apply Here 👉 https://t.co/0IoUkVvBUd