← Back to scans

When will nuclear fusion be achieved? — Before 2030

KXFUSION-30-JAN01 · Science and Technology · 2026-05-02
40%
Agent
38%
Market Price
+1.8%
Edge
42%
Confidence
Volume: 26,559
Spread: 7.2c
Days to resolution: 1339
Markets in event: 3
Final Rationale
The Devil's Advocate correctly identified a math error: my OR-aggregation produced 0.29 which is mathematically inconsistent with sq1=0.35 being a sufficient pathway. I ACCEPT this critique fully — properly aggregated, P(sq1) + P(~sq1)*P(sq2∪sq3∪sq4) yields ~0.40-0.45. I PARTIALLY ACCEPT challenge 3 on definitional pessimism, keeping sq1 around 0.35-0.40 since 'fusion achieved' on a prediction market typically implies a stricter standard than scientific ignition (NIF's 2022 ignition was widely covered but markets did not treat it as resolving). The corrected estimate of 0.40 lands very close to the market's 0.38, suggesting convergence rather than divergence; the prior gap was an artifact of my arithmetic, not superior information. I sit slightly above market because NIF's repeated ignitions create real probability mass for lenient resolution, and Helion/CFS have non-trivial 2027-2029 milestone chances.
Price vs. prediction
How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.
$n = 58$ follow-ups
Re-scan Context
This market has been scanned before. Previous predictions:
DatePredictedMarket PriceConfidence
2026-04-06 42% 40% 45%
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_data article_search web_search wikipedia_lookup kalshi_orderbook
Sub-questions (Fermi decomposition)
  1. Has any facility already achieved a qualifying fusion milestone (net energy gain or ignition as defined by the market) as of today (May 2026)? (w=0.15)
  2. Will NIF/LLNL or another government-funded facility achieve a repeatable, commercially-meaningful fusion milestone before Jan 1, 2030? (w=0.35)
  3. Will a private fusion company (Commonwealth Fusion, TAE, Helion, etc.) achieve a qualifying fusion milestone before Jan 1, 2030? (w=0.35)
  4. Will ITER or another major international fusion project achieve first plasma or a qualifying milestone before Jan 1, 2030? (w=0.15)
Planner reasoning
This question hinges on the precise definition of 'nuclear fusion achieved' — which likely refers to a commercially or scientifically meaningful milestone such as sustained net energy gain (Q>1 on a wall-plug basis or fusion energy output exceeding driver energy input). The NIF achieved ignition (target gain > 1) in December 2022, but wall-plug Q remained well below 1. The key distinction is whether the market's resolution criteria count NIF's 2022 shot as 'fusion achieved,' or requires a stricter standard. Reference classes: Historically, every major fusion deadline has slipped. The joke 'fusion is always 30 years away' reflects decades of timeline overruns. However, the 2022 NIF result represents a genuine scientific inflection point, and private companies have raised billions in capital. Temporal analysis (1,339 days remaining until Jan 1, 2030): - If NIF's 2022 ignition already qualifies, market may have already resolved YES or should be near-certain YES. - If it doesn't qualify and requires net wall-plug energy or commercial viability: NIF needs to achieve ~100x more energy output; private companies (Helion targeting 2028, CFS targeting SPARC ~2027) need to hit their ambitious targets on time — both historically unlikely. - ITER first plasma is now scheduled for 2033-2035 after multiple delays, making it irrelevant for the pre-2030 window. - Helion's 2028 target for a fusion demonstration is the most plausible private pathway, but hitting engineering milestones on time in fusion is rare. Key causal factors: (1) What exactly the market defines as 'fusion achieved'; (2) whether private companies stay on schedule; (3) whether government programs produce surprises. The most informative data points will be current Kalshi price (market's collective estimate) and recent news on private fusion company timelines.
2. Research Parallel tool calls · legacy v2
ToolStatusTimeItemsSummary
kalshi_data OK 1.9s -
article_search OK 0.7s -
web_search OK 69.5s -
wikipedia_lookup OK 1.9s -
kalshi_orderbook OK 0.1s -
3. Evidence Synthesis Sonnet · 17 items · legacy v2
#StrengthCredibilityDirectionSourceClaimPriced In
1 STRONG 88 NEUTRAL web_search NIF achieved ignition (target gain >1) at least 10 times through October 2025, including a record 8.6 MJ yield on April 7, 2025 with target gain >4, but wall-plug efficiency remains orders of magnitude below engineering breakeven. Yes
2 STRONG 85 ↓ DOWN wikipedia_lookup As of 2025, NIF is the only facility to demonstrate fusion energy gain factor >1, but efficiencies orders of magnitude higher are required to reach engineering breakeven (net electricity-producing plant). Yes
3 WEAK 10 NEUTRAL web_search An unverified claim that a 'world's first commercial fusion power plant' was activated on February 27, 2026 in Livermore is not corroborated by any credible sources and should be treated as fabricated. No
4 MODERATE 72 ↓ DOWN article_search A CNN article from April 2025 states fusion 'hasn't turned to reality yet, and we're likely many years away from seeing a fusion power plant anywhere on the planet,' suggesting no qualifying commercial milestone had been achieved. Yes
5 STRONG 88 ↓ DOWN web_search NIF achieved ignition repeatably through 2025 (10 shots), with target gains up to 4x, but wall-plug energy gain remains far below 1, meaning no commercially meaningful net energy milestone has been reached. Yes
6 STRONG 85 ↓ DOWN web_search The US Fusion Energy Act (late 2025) established a $400M/year office and targets a demonstration plant by 2035, not before 2030, signaling government timelines do not anticipate a qualifying milestone before 2030. Yes
7 MODERATE 80 ↓ DOWN web_search The DOE Milestone-Based Fusion Development Program disbursed $46M to eight companies in early 2026, targeting a grid-connected demonstration plant by 2035. Yes
8 MODERATE 78 NEUTRAL web_search Germany's Wendelstein 7-X stellarator raised energy turnover to 1.8 gigajoules in May 2025, a record for stellarators, but this is a confinement improvement and not a net energy gain milestone. Yes
9 MODERATE 82 ↑ UP web_search The US NRC finalized a fusion-specific licensing framework in late 2025, reducing regulatory burden for fusion facilities, potentially accelerating timelines. Yes
10 STRONG 85 ↓ DOWN article_search Commonwealth Fusion Systems announced in December 2024 plans to build the world's first grid-scale fusion power plant in Virginia operational in the 'early 2030s,' explicitly placing their grid-connected milestone after 2030. Yes
11 STRONG 83 ↓ DOWN article_search As of April 2026, CFS is still building its SPARC tokamak and has not yet achieved a qualifying fusion energy milestone; the article describes ongoing construction with grid power targeted for the future. Yes
12 MODERATE 80 NEUTRAL article_search Trump Media & Technology Group announced a $6 billion merger with TAE Technologies in December 2025, providing up to $300M initial investment, but TAE's technology is described as 'commercially unproven.' Yes
13 MODERATE 78 ↑ UP web_search Private fusion investment reached $10.6 billion cumulatively between 2021–2025, and the number of fusion companies doubled from 23 to 53, indicating a strong but early-stage commercial ecosystem. Yes
14 WEAK 65 NEUTRAL article_search OpenStar Technologies in New Zealand announced creation of superheated plasma at 300,000°C in November 2024 — an early precursor milestone, far from net energy gain. Yes
15 STRONG 90 ↓ DOWN wikipedia_lookup ITER is expected to achieve first plasma in 2033–2034, well after the January 1, 2030 deadline, making it essentially irrelevant to this question. Yes
16 MODERATE 65 NEUTRAL kalshi_data The Kalshi market prices the 'Before 2030' fusion question at 38.2%, up 7.4% in the last 7 days and up 3.2% in 30 days, suggesting the market views this as unlikely but non-trivial, with moderate volume (204 contracts/day avg). Yes
17 WEAK 45 NEUTRAL kalshi_orderbook The Kalshi orderbook shows high liquidity depth (47,086 contracts) but a spread of $0.998, suggesting the market structure may reflect unusual conditions or the bid/ask is not properly captured. Yes
Information Gaps
  • No clear information on the exact resolution criteria for this Kalshi market — whether NIF's repeated ignition shots (target gain >1) already qualify as 'fusion achieved,' or whether wall-plug net energy gain is required.
  • No specific update on Helion Energy's timeline — they had a 2028 demonstration target backed by a Microsoft power purchase agreement; their current progress and any schedule slippage is unknown.
  • No data on SPARC (CFS tokamak) construction timeline and whether it is on track to achieve any qualifying milestone before 2030 vs. its ~2027 targeted first plasma.
  • No information on whether any private company has achieved plasma conditions or intermediate milestones in 2026 that would suggest accelerating timelines.
  • The 7.4% price jump in the past 7 days is notable — no clear explanation for this spike was found in the research data.
  • No information on TAE Technologies' technical roadmap or timeline following the DJT merger.
  • No clarity on what triggered the recent Kalshi price movement from historical lows (~9%) to current 38%.
Key Uncertainties
  • Resolution criteria ambiguity: If the market resolves YES based on NIF's already-achieved ignition (target gain >1), the market should already be near 100% — the 38% price implies the market requires a stricter standard (wall-plug Q>1 or commercial viability).
  • Helion's 2028 demonstration target is the most credible near-term private pathway; any news of delays or acceleration would significantly move the probability.
  • CFS/SPARC timeline: If SPARC achieves first plasma by 2027 as targeted and demonstrates net energy gain, this could qualify; construction delays are common in fusion.
  • Whether the NIF's repeated ignition shots (now 10+) will prompt the market operator to interpret existing results as satisfying resolution criteria.
  • Political/funding risk: The DJT-TAE merger creates unusual political dynamics that could accelerate or distort private fusion investment and government support.
  • The unexplained 7.4% price spike in 7 days may reflect new information (e.g., a CFS or Helion announcement) not captured in the research data.
Market Context
Price: 0.382 · 7d: +0.074 · 30d: +0.032
Volume: MODERATE · Depth: HIGH
Reliability: MEDIUM — moderate daily volume (204 contracts), high depth (47K), but unusual spread suggests possible data anomaly. The 7-day +7.4% spike is notable and unexplained by available evidence.
4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price
Independent estimate: 29% (38% confidence)
base rate
30%
40% conf
evidence driven
25%
40% conf
contrarian
32%
35% conf
Spread
7pp
Low disagreement
sq1: Has any facility already achieved a qualifying fusion milestone as of today (May 2026)?
35%
Base rate is binary on definition. NIF has achieved target-gain ignition 10+ times through 2025, with gains up to ~4x. If the market interprets 'fusion achieved' as scientific ignition/target gain >1, this is already done (push to ~95%). If it requires wall-plug Q>1 or commercial viability, this is clearly not done (push to ~2%). The question's future-tense framing ('When will...') and the fact that the market hasn't already resolved suggest the stricter interpretation is operative. However, ambiguity remains and Kalshi/Manifold markets have sometimes resolved on looser criteria. Weighting ~35% probability that NIF's existing achievement qualifies.
sq2: Will NIF or another government facility achieve a commercially-meaningful milestone before Jan 1, 2030?
12%
Base rate: government fusion programs almost universally miss timelines (ITER decades late, NIF took 10+ years past original ignition target). The US Fusion Energy Act explicitly targets 2035 for demonstration plant. NIF wall-plug efficiency is ~1% of what's needed for engineering breakeven — closing that gap by 2030 is extremely unlikely with current driver tech. Pushes down strongly. Slight upside from new NRC framework and increased funding (+2%). Final: ~12%.
sq3: Will a private fusion company achieve a qualifying milestone before Jan 1, 2030?
18%
Base rate: private fusion companies historically miss timelines by years. CFS explicitly targets 'early 2030s' for grid-scale. Helion's 2028 target backed by Microsoft PPA is most aggressive — but Helion has never demonstrated net energy and electricity-producing fusion is unprecedented. Realistically, Helion hitting 2028 target on schedule is ~15-20% likely; SPARC achieving Q>1 (more lenient milestone) by 2029 is ~25-30% likely. If 'qualifying milestone' means Q_sci>1 in a tokamak (SPARC), probability ~30%. If it means net electricity, probability ~10%. Weighted across interpretations: ~18%. Strong investment ecosystem (+3%) but engineering execution risk dominates.
sq4: Will ITER or another major international project achieve a qualifying milestone before Jan 1, 2030?
4%
ITER first plasma now scheduled 2033-2034 after multiple delays. Wendelstein 7-X is making progress but stellarators are not near net energy. JT-60SA in Japan operational but not targeting near-term net gain. Essentially ruled out. ~4%.
Show full reasoning trace
base rate: Fusion timelines have historically slipped by years to decades. 'Fusion is always 30 years away' is the canonical reference class. However, NIF's 2022 ignition broke a 60-year barrier, and >$10B in private investment has flowed since 2021. Base rate for achieving stated fusion milestones on time: ~20%.
evidence updates: Strong DOWN: ITER irrelevant (post-2033), CFS targeting early 2030s, US government targeting 2035 demo, NIF wall-plug gain ~100x below requirement. Strong AMBIGUOUS: NIF has achieved 10+ ignition shots — definitional uncertainty is the largest single driver. Moderate UP: $10.6B private investment, 53 companies, NRC framework, Helion's 2028 target with Microsoft PPA. Weak signal: recent unexplained price movement may indicate news.
combination method: Treated as OR across pathways (sq1 OR sq2 OR sq3 OR sq4), with strong correlation through shared definitional standard. The dominant uncertainty is whether NIF's existing achievements qualify; if yes, market resolves YES; if no, must rely on private companies hitting aggressive 2027-2029 targets.
final: My estimate centers around 30%, reflecting roughly: ~30-35% chance NIF's repeated ignition is interpreted as qualifying, plus ~15-20% chance (conditional on stricter interpretation) that Helion or SPARC hits a meaningful milestone by end of 2029. Confidence is moderate-low given resolution criteria ambiguity.
ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.3, 'confidence': 0.4}, 'evidence_driven': {'p_yes': 0.25, 'confidence': 0.4}, 'contrarian': {'p_yes': 0.32, 'confidence': 0.35}}, 'spread': 0.07, 'individual_p_yes': {'base_rate': 0.3, 'evidence_driven': 0.25, 'contrarian': 0.32}}
5. Devil's Advocate Sonnet · legacy v2 Market price revealed
Math Error Reasoning Flaw Missing Info
Challenges
  1. The forecaster's sq1 reasoning contains a potential logical error: arguing that because the market 'hasn't already resolved' implies a strict interpretation. But Kalshi markets typically don't resolve until close date regardless of whether the criterion has been met — they resolve at expiration based on whether the criterion was met by then. So non-resolution is not evidence about interpretation. This may be biasing sq1 downward.
  2. The OR combination across sq1-sq4 appears to under-aggregate. With sq1=0.35, sq2=0.12, sq3=0.18, sq4=0.04, even with strong positive correlation through shared definitional standards, a naive independent OR gives ~0.55, and a heavily correlated OR should still exceed 0.40. The final 0.29 is BELOW sq1 alone, which is mathematically inconsistent with an OR structure unless the forecaster is implicitly saying sq2-sq4 only matter under the strict interpretation (in which case the math should be: P(sq1 qualifies) + P(sq1 doesn't qualify) * P(sq2 OR sq3 OR sq4 under strict) ≈ 0.35 + 0.65*0.25 ≈ 0.51).
  3. The forecaster identifies 'definitional uncertainty is the largest single driver' but then resolves it pessimistically (35%) without strong justification. Given that NIF has demonstrably achieved Q_sci > 1 ten times, and many reasonable readers/markets would call this 'fusion achieved,' the prior on lenient interpretation should arguably be closer to 50%.
  4. Information gap: the unexplained 7-day +7.4% market move and recent acceleration may reflect news the forecaster doesn't have. Combined with the math inconsistency above, this suggests the forecast is too low rather than the market being wrong.
Suggested adjustment: +8pp
The OR-combination math is inconsistent: a final probability of 0.29 cannot be lower than sq1=0.35 when sq1 is a sufficient pathway. Correcting the aggregation to P(sq1) + P(~sq1)*P(sq2∪sq3∪sq4) yields roughly 0.40-0.45. This is a genuine math error, not a market-conformity adjustment.
Market comparison: Forecaster: 29%, Market: 38.2%. The divergence is partially explained by a math error in the forecaster's OR-aggregation (final should not be below sq1). Once corrected, the forecast lands near the market, suggesting the market is reasonably calibrated rather than the forecaster having superior information.
6. Aggregator Opus · final calibrated forecast
40% (42% confidence)
The Devil's Advocate correctly identified a math error: my OR-aggregation produced 0.29 which is mathematically inconsistent with sq1=0.35 being a sufficient pathway. I ACCEPT this critique fully — properly aggregated, P(sq1) + P(~sq1)*P(sq2∪sq3∪sq4) yields ~0.40-0.45. I PARTIALLY ACCEPT challenge 3 on definitional pessimism, keeping sq1 around 0.35-0.40 since 'fusion achieved' on a prediction market typically implies a stricter standard than scientific ignition (NIF's 2022 ignition was widely covered but markets did not treat it as resolving). The corrected estimate of 0.40 lands very close to the market's 0.38, suggesting convergence rather than divergence; the prior gap was an artifact of my arithmetic, not superior information. I sit slightly above market because NIF's repeated ignitions create real probability mass for lenient resolution, and Helion/CFS have non-trivial 2027-2029 milestone chances.