Given the agreement of the US Dock Workers (International Longshoremen’s Association) to salary increases, both union and the port returned to the bargaining table on Jan. 15, 2025 to discuss automation and other issues. What’s the probability of a strike in Q1 2025.
Welcome to the Multi-AI Oracle. Today, it is 20%, down from yesterday’s 40%.
[Carolyn‘s note: I think it should be even lower, 5%. By contrast, I tried this question on four GenAI chatbots, each of these five times in a row. Gemini’s results ranged from “significant risk” to “very low.” ChatGPT ranged from 50% to “effectively zero.” DeepSeek ranged from 30%-40% to 25% –35%. Grok3 ranged from “moderate to high” to “less than 1%.” Another weirdness, just today: For the first time, the news feed sent totally irrelevant info to our Multi-AI Oracle. On the plus side, the Oracle’s fact-checking system detected these errors. My theory for so many weird GenAI activities lately: There Is a Disturbance in The Force. 🙁]
Obtaining outside newsfeed. News from various sources. Based on the most recent news articles, here’s a summary of the US dockworkers situation:
Most Recent Developments:
The US Trade Representative’s office has proposed new measures targeting Chinese vessels, suggesting charges of up to $1.5 million for Chinese-built vessels entering US ports. This development comes as tensions continue between the US and China over maritime trade issues. [Carolyn’s note: What the fork? Totally irrelevant.]
In New York state, corrections officers are currently on strike, with the National Guard being activated to help manage the situation. This is a developing situation affecting prison facilities across the state.
[Carolyn’s note: What the fork again?! Totally irrelevant.]
Labor Agreements and Resolutions:
A significant development occurred in January 2025 when the International Longshoremen’s Association (ILA), representing 45,000 dock workers on the US East and Gulf Coasts, reached a tentative agreement with employers on a new six-year contract. This deal successfully averted what could have been a major port strike. This was particularly noteworthy as it followed an earlier strike in October 2024, which was the first such action by longshoremen in nearly 50 years and affected 36 ports.
Overall Labor Activity Context:
2024 saw significant labor activism across the United States, with workers participating in 5,328,345 days of strikes throughout the year. This represented a substantial increase from 2023, which had already seen a 141% increase in strike activity compared to 2022, with 539,000 workers participating in 470 work stoppages.
Summarizing outside newsfeed:
“Recent news highlights include a proposal by the US Trade Representative’s office to impose hefty charges on Chinese-built vessels entering US ports, amid ongoing US-China trade tensions. In New York, corrections officers are on strike, prompting the activation of the National Guard to maintain prison operations across the state. Notably, in January 2025, the International Longshoremen’s Association reached a tentative six-year contract agreement, preventing a major port strike. This follows a significant increase in labor activism in 2024, with a dramatic rise in strike days compared to previous years, reflecting heightened labor unrest. These developments suggest ongoing labor and trade tensions within and beyond US borders.”
Fact-checking news summary:
1. **Fact**: The US Dock Workers (International Longshoremen’s Association) agreed to salary increases.
– **Check**: True. The summary states that a tentative six-year contract agreement, which likely includes salary increases, was reached.
– **Importance**: Critical to the resolution of the question.
– **Relevance**: True, as it directly impacts the probability of a strike.
2. **Fact**: The union and the port returned to the bargaining table on Jan. 15, 2025, to discuss automation and other issues.
– **Check**: Implicitly true. The exact date of returning to the bargaining table is not mentioned in the summary, but an agreement was reached in January 2025.
– **Importance**: Important to the resolution of the question.
– **Relevance**: True, as discussions on automation and other issues could influence labor actions.
3. **Fact**: The US Trade Representative’s office proposed imposing hefty charges on Chinese-built vessels entering US ports.
– **Check**: True, according to the summary.
– **Importance**: Less important to the resolution of the question.
– **Relevance**: True, though it indirectly relates to port operations, the direct impact on labor actions is not clear.
4. **Fact**: Corrections officers in New York are on strike, prompting the activation of the National Guard.
– **Check**: True, according to the summary.
– **Importance**: Unimportant to the resolution of the question.
– **Relevance**: False, as it does not relate directly to the port or dock workers’ negotiations or likelihood of a strike.
5. **Fact**: In January 2025, the International Longshoremen’s Association reached a tentative six-year contract agreement.
– **Check**: True, according to the summary.
– **Importance**: Critical to the resolution of the question.
– **Relevance**: True, as a tentative agreement directly impacts the probability of a strike.
6. **Fact**: There was a significant increase in labor activism in 2024, with a dramatic rise in strike days compared to previous years.
– **Check**: True, according to the summary.
– **Importance**: Important to the resolution of the question.
– **Relevance**: True, as this context can help assess the likelihood of future strikes.
Overall, most facts align with the information provided in the summary and are relevant to some degree, particularly those directly related to the dock workers’ agreement and labor conditions. The relevance of other facts varies based on their direct or indirect connection to the potential for a strike in Q1 2025.
Fact-checked summary:
The US Dock Workers (International Longshoremen’s Association) agreed to salary increases as part of a tentative six-year contract agreement reached in January 2025. This is critical to the resolution of the question about the probability of a strike in Q1 2025, as such agreements typically reduce the likelihood of labor unrest. Additionally, both the union and the port returned to the bargaining table to discuss automation and other issues, which could influence labor actions moving forward. Furthermore, there was a significant increase in labor activism in 2024, marked by a dramatic rise in strike days compared to previous years, which is important context for assessing the likelihood of future strikes. Overall, these factors, particularly the tentative agreement, are relevant and suggest that the probability of a strike in Q1 2025 might be reduced, though ongoing negotiations about automation could still pose challenges.
OpenAI gives a base rate of 0.15. The following were considered in order to produce this base rate:
In determining the base rate, I took into account historical instances of strikes following tentative agreements. Such agreements typically reduce the likelihood of strikes as they offer immediate concessions (such as salary increases) that satisfy many union demands. Additionally, although there was an increase in labor activism in 2024, the fact that a contract agreement has been reached this year suggests a stabilizing influence. However, issues like automation are still on the negotiating table, which could lead to unrest.
Some potential divergent considerations that might affect the base rate:
The key divergent factor that could increase the likelihood of a strike is the ongoing negotiation about automation. If these discussions turn contentious and fail, it could lead to a breakdown in negotiations and raise the probability of a strike. Additionally, overall increased labor activism could influence the union to leverage more aggressive tactics despite the contract agreement, especially if external factors such as economic downturns or significant policy changes occur.
The following chain of events are necessary for the question to resolve positively:
– The tentative six-year contract agreement reached in January 2025 must fail to be ratified by the union members, which has a low likelihood given the benefit of salary increases.
– The ongoing negotiations about automation and other issues must break down, leading to significant dissatisfaction among union members. This has a moderate likelihood as automation is a contentious issue
– There must be a decision to go on strike by the union, which typically requires significant unrest or dissatisfaction. Given the contract agreement, this has a low likelihood.
Querying Claude (AI predicts: 0.15 – confidence: 7)
Querying Mistral (AI predicts: 0.2 – confidence: 6)
Querying OpenAI (AI predicts: 0.2 – confidence: 6)
Question Type: Binary
Median from LLMs: 0.2
Base rate: 0.15 (from OpenAI)
SD: 0.02
Confidence: 6
Conf Mode: Normal
Mellers: 0.12
Reverse Mellers: 0.29
Theory of Mind: 0.3 (What did the LLMs think other LLMs predicted?)
Beta Distribution: 0.02
Close Type: B (B = cautious # closer to 50%; A/C = closer to extremes)
# LLM responses: 3
Model value: 0.2
We queried Claude, Mistral, OpenAI. A summary of what the responding AIs considered is as follows:
The consensus among the AIs suggests that the likelihood of a strike occurring in Q1 2025 is low due to the recent tentative six-year contract agreement that includes salary increases, a factor historically associated with reducing strike probability. The agreement satisfies key demands, thus pacifying union members to some extent. However, ongoing negotiations concerning the contentious issue of automation could lead to dissatisfaction if unresolved, albeit the immediate risk remains low. Increased labor activism in 2024 offers context but is offset by the stabilizing influence of the agreement. Nevertheless, the prediction could change if automation discussions falter significantly, if economic conditions worsen, or if unseen dissatisfaction arises. The base probability of a strike is estimated at 0.15, taking historical data and current circumstances into account.
How many seats will the Conservative Party win in Canada’s next federal parliamentary election?
Welcome to the Multi-AI Oracle. How many seats will the Conservative Party win in Canada’s next federal parliamentary election?
Model value:
* less than 172: 35%
* between 172 and 205: 40%
* between 206 and 240: 20%
* more than 240: 5%
[Carolyn’s note: More disturbance in The Force: Today, the AI-based news feed delivered nothing relevant to this question🙁]
Obtaining outside newsfeed. News from various sources:
Based on the provided news articles, here’s a summary of the key points about Canada’s upcoming election:
Past Multi-AI Oracle Forecasts
Ontario is holding a snap provincial election on February 27, 2024, about 15 months earlier than originally scheduled. According to recent polling, Premier Doug Ford’s Progressive Conservative (PC) Party maintains a strong lead heading into election day, though their support has seen a slight dip
The main party leaders are:
– Doug Ford (Progressive Conservatives)
– Bonnie Crombie (Liberals)
– Marit Stiles (NDP)
– Mike Schreiner (Green Party)
Key points from recent coverage:
– The latest Ipsos poll shows the PCs still significantly ahead of their rivals
– Opposition parties face an uphill battle against Ford’s PCs
– The election is being characterized not as a “change election” but as a “protect me from change election” amid economic uncertainties
– CBC has identified 10 key ridings to watch that could determine the final outcome
– There’s particular focus on urban/rural divide in some battleground ridings
The election was called early by Premier Ford in a snap election move, and voting takes place this Thursday. The Progressive Conservatives appear positioned to maintain power, though final results will depend on voter turnout and performance in key swing ridings.
Summarizing outside newsfeed:
“Ontario is set to hold a snap provincial election on February 27, 2024, called 15 months earlier than originally planned. Premier Doug Ford’s Progressive Conservative Party is leading in the polls, despite a slight decrease in support. The main opposition parties, led by Bonnie Crombie (Liberals), Marit Stiles (NDP), and Mike Schreiner (Green Party), face significant challenges in closing the gap. Recent coverage suggests that this election is more about maintaining the status quo amid economic uncertainties rather than seeking change. Key battlegrounds include urban and rural ridings, with CBC highlighting 10 particular ridings to watch. While the Progressive Conservatives are favored to retain power, outcomes will depend heavily on voter turnout and shifts in key swing areas. The reliability of these reports is uncertain, as the newsfeed may include unverified claims or biased perspectives.”
Fact-checking news summary:
1. **Fact:** Ontario is set to hold a snap provincial election on February 27, 2024, called 15 months earlier than originally planned.
– **Correctness:** True (assuming it’s stated accurately as part of recent accepted information).
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
2. **Fact:** Premier Doug Ford’s Progressive Conservative Party is leading in the polls.
– **Correctness:** True (as stated, although polls can vary).
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
3. **Fact:** The main opposition parties are led by Bonnie Crombie (Liberals), Marit Stiles (NDP), and Mike Schreiner (Green Party).
– **Correctness:** Dependent on date since leadership can change; currently correct for each party based on available data.
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
4. **Fact:** The main opposition parties face significant challenges in closing the gap.
– **Correctness:** True based on available polling and analysis.
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
5. **Fact:** This election is more about maintaining the status quo amid economic uncertainties than seeking change.
– **Correctness:** Subjective, but presented as an accurate general sentiment.
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
6. **Fact:** Key battlegrounds include urban and rural ridings, with CBC highlighting 10 particular ridings to watch.
– **Correctness:** Likely true based on typical election coverage styles and practices.
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
7. **Fact:** The Progressive Conservatives are favored to retain power.
– **Correctness:** True (given current political standings and commentary).
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
8. **Fact:** Outcomes will depend heavily on voter turnout and shifts in key swing areas.
– **Correctness:** True as a general principle in elections.
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
9. **Fact:** The reliability of these reports is uncertain, as the newsfeed may include unverified claims or biased perspectives.
– **Correctness:** True, as media often have varying levels of reliability.
– **Importance to Resolution:** Unimportant.
– **Relevance to Outcome:** False.
None of these facts are about the federal election, specifically about the number of seats that the Conservative Party of Canada might win. As such, they are unrelated and unimportant to resolving the question about the federal election.
Fact-checked summary:
Given the original question regarding how many seats the Conservative Party might win in Canada’s next federal parliamentary election, none of the fact-checked items directly address or are relevant to the question. They pertain instead to Ontario’s provincial election dynamics and are thus unimportant for predicting federal election outcomes. As such, these facts do not contribute to determining the number of seats the Conservative Party of Canada might secure in the upcoming federal election.
Querying Claude (AI predicts: [“less than 172”: 0.4, “between 172 and 205”: 0.35, “between 206 and 240”: 0.2, “more than 240”: 0.05] – confidence: 4)
Querying Mistral (AI predicts: [“less than 172”: 0.45, “between 172 and 205”: 0.35, “between 206 and 240”: 0.15, “more than 240”: 0.05] – confidence: 5)
Querying OpenAI (AI predicts: [“less than 172”: 0.2, “between 172 and 205”: 0.4, “between 206 and 240”: 0.3, “more than 240”: 0.1] – confidence: 6)
Question Type: Multiple Choice
Confidence: 5
# LLM responses: 3
Model value:
* less than 172: 40%
* between 172 and 205: 35%
* between 206 and 240: 20%
* more than 240: 5%
We queried Claude, Mistral, OpenAI. A summary of what the responding AIs considered is as follows:
The AI reasoning suggests that while the Conservative Party holds a moderate level of support based on historical seat counts and current polling data, significant gains are required to surpass 172 seats in the next Canadian federal election. Current polls indicate increased Conservative support, but not of an overwhelming nature. The output considers the first-past-the-post electoral system and regional support distribution as factors influencing seat conversion efficacy. However, these predictions are subject to potential inaccuracies due to the volatile political climate, unreliability of polling data, and unforeseen political developments such as scandals, major events, or strategic changes, which could significantly alter voter opinion before the election.
Runtime for both questions on this page: 181 seconds.
Guide to our Multi-AI Oracle’s Reports
Median: We collect the output values from the LLMs in an array. We then take the median value. This is between 0 and 1 (i.e. 0.15 = 15%).
Base rate: Currently we take OpenAI’s value as the base rate. We are working on a better value using a different prompt that takes into account process. The base rate helps to determine if the median makes sense.
SD: The standard deviation between the median and the base rate.
Confidence: We query each of the LLMs on how confident they are of their predictions (between 0 and 10) and take a median. Because the LLMs tend to be overconfident, we take anything lower than 6 as being low confidence. This factors into the overall model.
Conf Mode: Based on the confidence value. >=9 is high confidence. Below 6 is low confidence (this is also triggered by an exceptionally high SD).
Mellers: This refers to Barbara Mellers, specifically a paper she wrote that includes a formula for moving values towards an extreme (i.e. 0 or 1).
Reverse Mellers: This uses the formula from above, but with a sub-1 coefficient to move the values closer to 50%.
Theory of Mind: We ask the LLMs what they think other LLMs would predict. We hope that this makes them consider the questions more deeply.
Beta Distribution: Currently unused, but possibly of interest. This is based on the median, the base rate and the SD.
Close Type: We noticed that the appropriate base case for some questions is closer to the extremes, while others are closer to 50%. When we have a low confidence value, this helps us to determine whether to extremize or de-extremize the value. ‘A’ implies closer to zero. ‘B’ implies closer to 50%. ‘C’ implies closer to 100%.
# LLM Responses: The count of LLMs that responded. We query 5 currently, but they have a relatively high API failure rate.
Model value: For normal confidence mode, this is the median. For high confidence, this is the Mellers value. For low confidence, this is either the Mellers or Reverse Mellers (depending on the Close Type).
