Put AI to work where it pays back first.
You do not need more AI news. You need to know which work to hand to AI this quarter, with which tools, and what your competitors already did. Spezia reads the market every day and turns it into decisions for a business like yours.
What you get done
Decide where to deploy first
ProThe opportunity map ranks the operations where AI is paying back now, with the evidence behind each call.
Opportunity map →Choose tools without a six-week evaluation
FreeThe tooling guide names the best model and tool for each kind of work this month, what it costs and what to avoid.
Tooling guide →Stay ahead in ten minutes a day
FreeYour front page ranks the day against your goals, and the weekly brief is ready to forward to your team.
Your front page →Get a straight answer
FreeAsk what a release, a price change or a competitor's move means for you, and get the sources with it.
Ask Spezia →
From this week's map
Real output, read from the data Spezia works from today. Every line links to its source.
With hyperscalers and labs committing hundreds of billions to data-center capacity while voters and new governors throttle permits and power, the biggest near-term return is in the unglamorous bottlenecks that let committed AI capital actually energize and run: power procurement, grid interconnection, and on-site generation.
- Businesses to build12 monthsmedium confidence
Entrepreneurs should build on-site and bridging power businesses (used turbines, converted jet engines, behind-the-meter generation) for data-center developers, since projected gas turbine orders of 129 GW exceed global production capacity of 60-70 GW and buyers are already scouring for second-hand turbines.
Basis: Source reports 129 GW projected turbine orders against 60-70 GW global production capacity and buyers turning to used turbines.
- Put AI to work in your operations3 monthsmedium confidence
Operators should deploy coding and research agents internally now, because METR found AI may have doubled Anthropic's research pace (acceleration factor 1.97) and more than 80% of merged code was authored by Claude, and OpenAI reports 3.1 agent workdays per human workday.
Basis: Source reports a 1.97 research acceleration factor, over 80% Claude-authored merged code, and 3.1 agent workdays per human workday.
- Where capital is going next12 monthsmedium confidence
Investors should favor data-center conversion platforms with contracted power, because Core Scientific controls 2.6 GW gross and 1.7 GW leasable capacity across 11 data centers and TeraWulf expanded Muskie to 1,000 MW contracted power.
Basis: Sources and report 2.6 GW gross/1.7 GW leasable capacity and 1,000 MW contracted power.
Sources:Core Scientific controls 2.6 GW of power across 11 data centers and is converting sites to AI and HPC workloads.TeraWulf expands Muskie Data Campus to 1,000 MW of contracted Kentucky power for AI/HPC.
The full set of calls, each with its sources, is inside.
Continue free →Free account, no card. A free account previews Opportunity map before you choose a plan.
Projections, not facts. Each states its basis and horizon, and its confidence is the lower of two independent checks.
Questions people like you ask first
Press one to run it. Each answer cites its sources.
Pro
$199 a month, or $166 a month billed annually.
Pro adds the opportunity map and 150 questions a month, which is what a weekly decision rhythm needs. Start on Explorer and move up when you hit the limit.
Compare every plan →- Everything in Explorer, with every Pro page open
- Full AI Intel: every story, all ten areas, 30 days, every source
- The weekly outlook: projections checked against their sources
- Opportunity map: what to build, where to deploy AI, where capital is going
- Money across AI: rounds, debt, acquisitions and spending plans
- Every AI lab, company and investor, ranked
- Interviews and talks, distilled into points with timestamps and quotes
- 150 questions a month to Ask Spezia
- Spezia inside Claude Code, Codex and Cursor: 1,500 calls a day
- 3 days a month inside the U.S. data center build-out
Questions
I am not technical. Is this for me?
Yes. Everything is written in plain language and ends with what to do. When a setup step is technical, Spezia gives it as numbered steps you can hand to someone.
Can my team use it?
Team includes five seats. Larger groups are set up on Enterprise.
Research written for operators and executives
- Power and the gridAnalysis56%of 1 GW+ projects are stuck mainly on power
Gigawatt data centers are 2.7 times as likely to be stuck on power
Power is the main limit on 56% of U.S. data center projects of 1 GW or more, against 21% of small ones. The data, and what utilities are doing.
- The AI build-outAnalysis285 GWplanned across 294 U.S. projects
The U.S. AI data center pipeline: 285 GW across 294 projects
294 U.S. data center projects plan 285 GW. See how much is under construction, how it splits by size, and what is holding it back.
- The AI build-outAnalysis57 GWplanned in Texas, 20% of the U.S. pipeline
Which states are building the most AI data center capacity? Texas leads
Texas leads U.S. AI data center capacity at 57.0 GW, ahead of Utah and Wyoming. Projects and capacity for all 41 states.
Spezia for other readers
- Founders
Build where the money and the demand are heading.
People building or funding a company on AI
- Investors
Track the capital in AI with the source next to every number.
Venture, growth, public-market and family-office investors
- Chips, power and infrastructure
Know which data center projects are real, and who is paying.
Developers, utilities, suppliers, contractors and the trades
- Engineers and career builders
Learn the AI skills and tools that pay, in the right order.
Engineers, skilled workers and students