Scenario wins: preseen (66) lewinke-thinking-bot* (59) smingers-bot (20) Panshul42 (19) Mantic (14) hayek-bot (14)
Question: net change in entities on the BIS Entity List published in the Federal Register between June 1 and August 12, 2026. Both forecasters note a clear downward trend in summer net additions (about 105–107 in 2022, 47 in 2023, 9–14 in 2024, and 0 in 2025) and point to a key policy change: a one‑year suspension of the “Affiliates Rule” after a US‑China trade deal on Nov 10, 2025, and a subsequent pause in Federal Register Entity List rules since then. One model (gemini3pro_1) places a high probability (66%) on a net change of 0, arguing the ongoing pause will likely persist through the short 73‑day window, with small chances of additions (targeting Russia/Iran) and very low odds of removals. The other (gemini3pro_2) is more cautious: about a 34% chance of 0, ~40% for small–medium additions (1–30 entities) as the US pursues evasion networks, and a tail risk of a large (>60) batch if geopolitics shift; it also notes removals are possible but historically rare. In short, both forecast low activity relative to 2022–23, but they disagree on how strongly to weight a zero outcome versus modest additions, with the main drivers being the Affiliates Rule suspension and geopolitical developments.
Forecast: Probability distribution:
We are forecasting an integer net change in Entity List entities from BIS rules published in the Federal Register during 2026-06-01 to 2026-08-12.
Across analysts, the key disagreement is whether the late-2025 to mid-2026 publication freeze persists through the window (implying N=0), or whether BIS “unpauses” with one or more catch-up batches (implying tens to potentially >100 net additions). Forecasters 3 and 4 weight the freeze/truce dynamics heavily (median 0), while forecasters 1 and 2 lean more on pre-freeze base rates and expect a resumption within the 10.5-week window (medians 25–38).
I synthesize these by treating the outcome as a mixture of regimes:
Removals are possible but structurally harder (unanimity) and historically sparse, so net negatives are a low-probability left tail rather than a central scenario.
Given the strong evidence of an ongoing multi-month pause but also the non-trivial chance of an administrative/geopolitical “restart” within a ~2.5 month window, I keep the median essentially at 0 while allowing a wide, right-skewed tail. I express about ~5% probability mass beyond the open upper bound by placing the 95th percentile above 169.9975 (and not near the bound).
An amateur forecast would commonly do one of two things: (a) extrapolate from 2025 batch sizes and predict a fairly tight 20–60 range, ignoring the unusual multi-month publication gap; or (b) predict 0 with near-certainty, underpricing the possibility of a sudden restart/backlog release. My forecast explicitly models a pause-vs-resumption mixture, keeps a realistic (but small) left tail for net removals, and uses an explicit past-upper percentile to represent the chance of an unusually large summer catch-up. Confidence in improvement is moderate: the dominant uncertainty is political/administrative timing rather than a slowly varying trend.
Forecast rationale (numeric):
— Iteration 1 — Across the forecasts, the dominant view is that the net change in BIS Entity List entities over the June 1–August 12, 2026 window is most likely to be 0 or very close to 0, with the main uncertainty coming from whether the current publication pause continues.
The shared reasoning is that the most plausible outcome is no net change, but the forecast is bimodal: either continued inactivity or a sudden, sizable batch of additions if BIS resumes action. The downside tail from removals exists but is comparatively small.
— Iteration 2 — Across the forecasts, the main reasoning is that the U.S. BIS Entity List is likely to remain inactive or nearly inactive during the June 1–August 12, 2026 window because of a recently reaffirmed multi-month pause tied to the broader U.S.–China de-escalation/truce environment. That pause is treated as the dominant factor, making 0 net change the most likely outcome.
Key patterns in the reasoning:
Strong baseline expectation of no action:
All models emphasize the six-month lull in Entity List rules as the central signal. This creates a sizable probability that no Federal Register rule appears in the window.
If action resumes, it is likely to be lumpy and addition-heavy:
Historical behavior suggests BIS tends to publish batch additions rather than small, steady changes. If the pause ends, additions are expected to come in groups, often totaling tens of entities, with some models allowing for much larger batches.
Right-skewed distribution:
The forecasts consistently describe the outcome space as heavily skewed toward positive net additions if any rule appears, with a large mass at zero and a long upper tail for escalation scenarios.
Removals are possible but much less likely:
Net removals are treated as a thin left tail outcome—generally limited in size and associated with diplomatic concessions or narrow delisting actions.
Backlog and broader sanctions policy matter:
Some rationales note that paused or delayed activity in related sanctions areas, especially involving Russia and Iran, could produce a catch-up batch of listings. Conversely, a truce breakdown or major geopolitical shift could drive larger-than-normal additions.
Where the forecasts differ is mostly in how much upside risk to assign:
Overall, the shared view is: no net change is the single most likely outcome, but the distribution is asymmetric, with meaningful risk of a positive batch-addition surprise if BIS resumes issuing Entity List rules.
— Iteration 3 — Overall, the forecasts converge on a strongly right-skewed distribution with a median net change of 0 for the June 1–August 12, 2026 window.
Recent publication pause is the dominant factor.
All rationales emphasize that BIS has been in an unusually long lull in Entity List rulemaking, making no Federal Register rule in the window the most likely outcome.
Recent U.S.–China diplomacy is seen as reinforcing the pause.
The mid-May 2026 Trump-Xi engagement, along with the broader post-summit truce framework, is interpreted as lowering the near-term chance of new listings.
If the pause ends, changes would likely come in batches.
Rather than small incremental updates, Entity List actions are expected to arrive as clustered rules with multiple entities, especially if a backlog of Russia, Iran, or China-related designations is released.
Removals are treated as rare and constrained.
Negative net change is considered unlikely because delistings require a higher procedural threshold and are historically uncommon, so the left tail is thin.
Summary of Forecasting Rationales
Forecasters agree that the net change in U.S. BIS Entity List additions during the summer 2026 window hinges on the tension between an ongoing diplomatic freeze and mounting political and international pressure to resume routine designations.
The Diplomatic Pause and Policy Shift The primary factor anchoring the forecasts is the unprecedented six-month pause in Entity List additions that began in November 2025. Forecasters note this freeze is directly tied to a “tactical truce” between the U.S. and China stemming from the Busan summit, alongside the formal suspension of the “Affiliates Rule.” To avoid jeopardizing fragile economic agreements, the administration has largely halted Commerce Department additions. Instead, enforcement has shifted toward alternative mechanisms, such as the Treasury Department’s Specially Designated Nationals (SDN) list, the Unverified List (UVL), and enhanced “Know Your Customer” guidelines. This dynamic strongly supports the possibility that the administration maintains the freeze, resulting in a stagnant list.
Catalysts for Resumption and Backlog Risks Despite the overarching truce, forecasters highlight several imminent catalysts that could break the pause and trigger a wave of new additions:
The Rarity of Removals Forecasters are unanimous in constraining the lower bound of their estimates due to the structural mechanics of the ERC. Removing an entity from the list requires a unanimous interagency vote, an exceptionally rare occurrence. Consequently, significant negative net changes are highly improbable, limiting the downside strictly to isolated administrative clean-ups and ensuring the distribution of outcomes is heavily skewed toward either a maintained pause or a sudden positive spike.
Agent Forecast (discrete, confidence: low)
Base Rate: Biden 2024: 520 entities/year → ~104 entities per 73-day window. Trump 2nd term: ~140/year equivalent with 81-day and 171-day gaps; current 2026 pace: 0 entities in 186+ days. Conditional on any resumption, historical Trump 2nd-term batch sizes: 13, 26, 32, 82 entities.
Initial Reasoning:
Key Uncertainties:
Evidence Sources (9):
Grounding Facts (passed to reviewers):
Draft: 0: 35.0%, 171+: 1.8%, 32: 1.2%, 26: 1.2%, -1: 1.0%
Reviewer Panel:
Reviewer (reasoning)
The forecast’s linchpin—that the Affiliates Rule suspension effectively freezes all new listings—misreads the rule’s scope and overweights a single base rate gap. It also leans on outdated Biden-era listing volumes rather than Trump-term episodic patterns, and fails to consider ongoing Russia/Ukraine and Iran sanctions momentum. Finally, the elaborate probability mass on dozens of nonzero outcomes is unsupported by any discrete signals of an imminent rule.
Objections:
Reviewer (calibration)
The forecast identifies ‘strong headwinds’ against Entity List changes due to ongoing diplomatic accommodation and the Affiliates Rule suspension, yet assigns only a 35% probability to zero net changes. This appears contradictory, as the described policy context strongly suggests a continuation of the current pause. The distribution of 62.5% across a wide range of positive outcomes seems to underestimate the impact of the active policy of détente.
Objections:
Reviewer (fact-checker)
2 unsupported claim(s). Verdict: minor_issues
Unsupported claims:
Adjustment Rationale:
Adjusted P(N=0) from 0.35 → 0.58 in response to the calibration reviewer, who correctly identified that my draft’s qualitative reasoning (unprecedented 186+ day pause, US-China diplomatic framework, BIS’s 2026 enforcement-only agenda, explicit Affiliates Rule suspension through Nov 2026) implied stronger weight on a continued pause than 35% reflected. The ensemble signal (68%) and both reviewers independently pulled toward a higher zero-weight. I stopped short of 0.68 to preserve mass for Trump’s well-documented episodic burst pattern, where long pauses (81 and 171 days in 2025) have historically been followed by typical rule sizes (13, 26, 32, 82 entities). The reasoning reviewer’s technical point — that the Affiliates Rule suspension only alters deeming provisions and does not legally bar new designations — is correct, but the constraint operating here is diplomatic/political, not legal, so it does not change the directional adjustment. The fact-checker flagged two of my claims (Trump-Xi May 2026 framework intact, Russia/Iran ‘nominally active’) as inadequately sourced; I am treating both as inference rather than evidence and have softened my reliance on them. Concentrated remaining positive mass at historically observed Trump-era batch sizes (peaks at 26, 32, 82) and trimmed the long right tail.
forecast_sdk | 531.6s | research $0.8000 + reviewers $0.0093 + reconcile $0.2456 = $1.0550
I forecast N = 0 as the modal outcome, with a 56% probability of no qualifying rule and a mean net change near +13 entities.
The question counts only BIS final or interim final rules that amend Supplement No. 4 to 15 CFR part 744, published in the Federal Register from June 1 through August 12, 2026, using the “entities added” minus “entities removed” counts in each rule summary. Address-only entries, aliases, typographical changes, and broad policy rules that do not add or remove named Entity List entities score 0.
The current state is quiet. I found no 2026 BIS Entity List amendment rule through May 17, 2026 in the Industry and Security Bureau Federal Register record; the latest qualifying Entity List amendment I found was the November 12, 2025 rule removing one China entity and removing six aliases from another entry (Federal Register, 2025-19858). The same Federal Register issue stayed the Affiliates Rule through November 9, 2026, while saying BIS would continue to evaluate national-security and foreign-policy interests related to non-listed affiliates (Federal Register, 2025-19846).
The historical base rate is high. I used the prior June 1-August 12 windows from 2017 through 2025, with unit = net distinct entities, coverage window = Federal Register publication dates in each annual 73-day window, sample size = 9 annual windows, and vintage = public Federal Register text available on May 17, 2026. The mean is +35.9 entities, the median is +44, and 8 of 9 windows had a positive net change. I count the June 21, 2023 correcting amendment as +1 because its summary says it added an inadvertently omitted entity; if the resolver treats that correction as 0, the historical mean falls by only 0.1 entity.
| Year | Qualifying Entity List activity in June 1-August 12 window | Net N |
|---|---|---|
| 2017 | One rule adding 10 entities (Federal Register, 2017-13167) | +10 |
| 2018 | One rule adding 44 entities (Federal Register, 2018-16474) | +44 |
| 2019 | One rule adding 5 entities (Federal Register, 2019-13245) | +5 |
| 2020 | Three rules adding 9, 24, and 11 entities (Federal Register, 2020-10868, 2020-10869, 2020-15827) | +44 |
| 2021 | Six rules; 57 additions and 3 removals (Federal Register, 2021-11304, 2021-12751, 2021-13395, 2021-14367, 2021-14656, 2021-15362) | +54 |
| 2022 | Two rules; 107 additions and 2 entity removals; the June 30 rule also removed one address, which I do not count (Federal Register, 2022-12144, 2022-14069) | +105 |
| 2023 | June 14 additions/removal, June 21 correcting amendment adding one omitted entity, and July 19 additions (Federal Register, 2023-12726, 2023-13196, 2023-15343) | +47 |
| 2024 | Three rules adding 5 entities plus 8 addresses, then 3 entities, then 6 entities; addresses do not count (Federal Register, 2024-13148, 2024-13695, 2024-14635) | +14 |
| 2025 | No qualifying June 1-August 12 rule; later rules added 32 entities on September 16, added 26 entities and 3 addresses on October 9, and removed 1 entity on November 12 (Federal Register, 2025-17893, 2025-19508, 2025-19858) | 0 |
That base rate cannot be used raw. The latest no-action spell is long: from the November 12, 2025 removal rule to May 17, 2026 is 186 days with no qualifying Entity List amendment that I found (Federal Register, 2025-19858; BIS Federal Register archive). The previous long quiet spell, from the March 28, 2025 additions to the September 16, 2025 additions, was 172 days and already produced a zero in the 2025 target-season window (Federal Register, 2025-05426; Federal Register, 2025-17893). I read this as a real policy and process signal, not just ordinary calendar noise.
The signal is not strong enough to make zero overwhelming. The Entity List is not only a China tool. Recent rules covered China, Iran, Pakistan, South Africa, Turkey, the UAE, India, Singapore, Taiwan, and Russia-related diversion networks (Federal Register, 2025-05426; Federal Register, 2025-17893; Federal Register, 2025-19508). Also, BIS rules state that the End-User Review Committee adds entries by majority vote, but removals and modifications require unanimity, so negative outcomes are rare and usually small (Federal Register, 2022-14069).
My final distribution puts 56% on N = 0, 2.5% on a negative net, 41.2% on positive values from 1 to 169, and 0.3% above 169.9975. The mean is about +13.5 if the upper-tail bin is represented by 180. The 90th percentile is in the low 40s. The positive mass is centered around single-rule batches in the 10-45 range, with smaller mass on 70-120 entity batches because those have occurred, but are less likely during the current truce-like period.
The simple historical answer is too high. Summer Entity List activity was common in 2017-2024, but the 2025 summer window was already zero and the current drought is longer than the 2025 drought. A model that just says “8 of the last 9 windows were positive” misses the regime break.
The simple political answer is also too low. The Affiliates Rule suspension reduces one source of China-linked escalation, but it did not suspend ordinary named Entity List additions, and the rule itself says BIS will continue evaluating national-security and foreign-policy interests during the stay (Federal Register, 2025-19846). If the pause breaks, the result is more likely to be a positive batch than a negative rule, because removals are procedurally harder and historically much smaller.
The main missing variable is the closed ERC pipeline. Federal Register publication is visible only at the end of the process, and Entity List rules often arrive as final rules with little public warning. A prepared Russia, Iran, cyber, military-end-use, or transshipment batch could appear inside the window without a public lead time.
The historical sample is small: 9 comparable annual windows and a few dozen rules. Classification also has edge cases. Corrections and address-only entries matter because the resolution uses entity counts, not entry counts. I handled the 2023 correction as an entity addition because its summary says it added an omitted entity, and I excluded address-only additions such as the 2024 high-diversion-risk addresses and the 2025 three addresses. That choice moves the historical mean by far less than the policy uncertainty moves the forecast.
Median: 3.95 entities
Dominant scenario is continued pause (44–65% probability). The 6+ month publication gap since November 2025 is historically unprecedented and reflects a formal diplomatic constraint tied to the Affiliates Rule suspension through November 9, 2026. The May 2026 Trump–Xi summit produced no explicit release of this constraint, leaving the pause as the baseline expectation.
Meaningful tail probability for non-China rules (20–28% probability). Iran, Pakistan, and Russia-related entity listings remain active in the ERC pipeline and are not explicitly covered by China-focused diplomatic agreements. A single modest batch targeting these regions could add 10–40 entities without breaking the China truce.
Lower but non-negligible risk of China rule resumption (12–20% probability). The May summit’s ambiguity—no new commitment but also no explicit release of constraints—leaves room for the administration to interpret silence as permission. Z2Data’s April 2026 analysis noted the administration may use the Entity List as leverage through Q3 2026, suggesting late-summer action is possible.
Structural asymmetry favors additions over removals. ERC voting requires a majority for additions but unanimity for removals. Net removals are historically rare (1 entity removed in all of 2025) and would require a deliberate diplomatic concession, making the left tail thin.
Batch-size distribution is right-skewed. Historical precedent shows batches of 13–70 entities when BIS publishes. The weighted average of scenarios (combining 55–65% at zero with 35–45% distributed across 10–85 entity ranges) produces a median near 0 but a mean of 15–30, consistent with the forecast median of 3.95.
Net change in entities on the U.S. BIS Entity List via Federal Register rules published between June 1 and August 12, 2026?
Key figures
Historical context
Historically, the BIS Entity List has been a dynamic tool of U.S. foreign policy, often characterized by ‘batch’ additions rather than a steady trickle. In early 2025, the list grew significantly with rules published on March 28 (82 entities total), September 16 (32 entities), and October 9 (26 entities). These actions were largely focused on China, Russia, Iran, and Pakistan. However, the regulatory environment shifted dramatically following the October 30, 2025, Busan summit. On November 12, 2025, the BIS published a final rule that both removed one entity and stayed the ‘Affiliates Rule’ expansion. Since that date, there has been a complete cessation of Federal Register rules amending the list. This six-month gap is highly unusual compared to the activity levels seen in early 2025, reflecting a ‘tactical truce’ that has held through the most recent Trump-Xi summit in May 2026. During this pause, the BIS has focused on civil penalties and settlements with existing firms (like Applied Materials and Teledyne FLIR) rather than adding new names to the list.
Tailwinds
Headwinds
Detailed reasoning
My analysis for the net change in the U.S. BIS Entity List during the June 1 to August 12, 2026, window is primarily informed by the current period of unprecedented regulatory inactivity and the high-level diplomatic “truce” between the United States and China. As of May 17, 2026, the Bureau of Industry and Security (BIS) has not published a rule amending the Entity List in over six months, with the last action occurring on November 12, 2025. This pause is not accidental; it is a direct consequence of the October 30, 2025, Busan summit, which established a tactical de-escalation framework. A central pillar of this truce is the formal stay of the “Affiliates Rule” expansion, which is effective through November 9, 2026. This rule would have automatically added hundreds of majority-owned subsidiaries to the list, but its suspension effectively freezes the most significant source of list growth.
Furthermore, the May 14–15, 2026, Beijing summit between Presidents Trump and Xi—occurring just two weeks before the start of the forecast window—was characterized by analysts as a “stalemate” that preserved the existing truce. While the U.S. Trade Representative has hinted that this truce might not be extended past its November expiration, there is little evidence to suggest a major policy reversal will occur during the summer of 2026. Because the question resolves to 0 if no qualifying rule is published, and the current administrative silence has lasted half a year, a net change of zero is the single most likely result.
However, I must account for the “batch” nature of BIS administrative actions. Historically, when the End-User Review Committee (ERC) decides to add entities, it does so in clusters. In 2025, rules were published in batches of 70, 32, and 26 entities. While the China-specific truce is holding, BIS still maintains active programs regarding Russia, Iran, and other regions. The Trump administration has requested a massive 92% budget increase for BIS in FY 2027 ($450 million), intending to fund 290 new export enforcement agents. This buildup of capacity suggests that even if the “Affiliates Rule” remains stayed, the agency is preparing for a more aggressive enforcement posture. If the current “silence” is broken by a non-China-related rule or a tactical batch meant to signal strength before the November midterms, the net change would likely jump immediately to the 20–40 entity range. Removals are considered highly unlikely to offset additions, as removals require a unanimous vote from the ERC (Commerce, State, Defense, Energy, and Treasury), whereas additions require only a simple majority. My prediction weights the “status quo” (N=0) most heavily but includes a tail of positive “chunky” outcomes to reflect these administrative patterns.
Key uncertainties
Conclusion