The International Longshoremen’s Association 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, 35%. [Jeremy’s note: News feed isn’t picking up that the union leadership have tentatively ratified the agreement. I need to do some work on the bot....]
Obtaining outside newsfeed. News from various sources. Based on the most recent news articles, here’s a summary of the US dockworkers situation:
The most significant recent event was a major dock workers’ strike that began on October 1, 2024, involving 45,000 International Longshoremen’s Association (ILA) workers across 36 ports on the US East and Gulf Coasts. This was the first such strike in nearly 50 years. However, the strike was short-lived, as workers and the United States Maritime Alliance (USMX) reached a deal on October 3, 2024, leading to an immediate return to work. [Carolyn’s note: Not exactly true. It was a tentative agreement. Negotiations resumed Jan. 15, 2025. See also above, Jeremy reports a tentative new agreement.]
More recently, there have been localized port disruptions, such as a strike at Hutchison Port Delta II in Rotterdam that has caused delays, with Maersk warning customers about ongoing slowdowns even after operations resumed. The impact of these port worker actions continues to raise concerns about supply chain disruptions, potential inflation, and shortages of various goods, particularly agricultural exports.
In related labor news, other sectors are experiencing similar tensions – workers at an Indianapolis plant making military jet engines are facing a contract deadline, and there’s potential strike action brewing among 30,000 City of Toronto workers, with 90% voting in favor of a strike mandate. Additionally, Costco workers across the United States have voted to authorize a nationwide strike over stalled contract negotiations and demands for better wages.
Summarizing outside newsfeed:
“Recent news highlights significant labor disputes involving dock workers and potential strikes in other sectors, though information may include rumors and propaganda. A notable event was a large-scale strike by 45,000 International Longshoremen’s Association workers across the US East and Gulf Coasts, starting on October 1, 2024, marking the first event of this magnitude in nearly five decades. A quick resolution with a deal on October 3, 2024, ended the strike. However, localized disruptions continue, with a strike at Hutchison Port Delta II in Rotterdam leading to delays and warnings from Maersk about ongoing slowdowns. Concerns are rising about potential impacts on the supply chain, inflation, and goods shortages. Similarly, labor tensions are seen in other sectors, such as military jet engine manufacturing in Indianapolis, looming potential actions in Toronto’s municipal workforce, and a nationwide strike authorization by Costco workers over stalled negotiations.”
Fact-checking news summary:
Based on the provided summary, let’s identify the specific facts mentioned, evaluate their accuracy, importance, and relevance to the probability of a dock workers’ strike in Q1 2025:
1. **Fact:** Dock workers have agreed to salary increases.
– **Check:** Not explicitly mentioned in the summary. Assumed from the question context. True if based on agreement.
– **Importance:** Critical (this agreement is a part of the basis for returning to bargaining).
– **Relevance:** True (relevant to the question about potential strike probabilities).
2. **Fact:** Dock workers’ union and the port returned to the bargaining table on January 15, 2025.
– **Check:** Not directly stated in the summary.
– **Importance:** Important (details about ongoing negotiations affect strike likelihood).
– **Relevance:** True (negotiations directly impact the probability of a strike).
3. **Fact:** Large-scale strike by 45,000 International Longshoremen’s Association workers on October 1, 2024.
– **Check:** Internally consistent and specific within the summary.
– **Importance:** Important (indicates recent tensions and resolution patterns).
– **Relevance:** True (recent actions by dock workers inform potential future actions).
4. **Fact:** Strike ended with a deal on October 3, 2024. [Carolyn’s note: Misleading. it was a tentaticve deal with negotiations slated to resume Jan. 15, 2025]
– **Check:** Internally consistent within the summary.
– **Importance:** Important (demonstrates the capacity for resolution).
– **Relevance:** True (implies capability and willingness to resolve disputes quickly).
5. **Fact:** Localized disruptions continue, notably a strike at Hutchison Port Delta II in Rotterdam.
– **Check:** Consistent with the summary.
– **Importance:** Lesser (specific location not related to US strikes).
– **Relevance:** False (not directly relevant to US dock workers).
6. **Fact:** Maersk has issued warnings about ongoing slowdowns.
– **Check:** Consistent with the summary.
– **Importance:** Lesser (indicates industry-wide concerns but indirectly related).
– **Relevance:** False (specific to logistics impacts, not US dock worker negotiations).
7. **Fact:** Concerns about potential impacts on supply chain, inflation, and goods shortages.
– **Check:** Consistent with summary commentary.
– **Importance:** Important (broad economic context affecting strike probabilities).
– **Relevance:** True (economic impacts can influence negotiation pressures).
8. **Fact:** Labor tensions in other sectors, such as military jet engine manufacturing in Indianapolis.
– **Check:** Consistent with the summary.
– **Importance:** Unimportant (specific sectoral details not directly related to dock workers).
– **Relevance:** False (differs from the dock worker context).
9. **Fact:** Looming potential actions in Toronto’s municipal workforce.
– **Check:** Consistent with the summary.
– **Importance:** Unimportant (cross-national relevance is limited).
– **Relevance:** False (distinct from US dock worker industrial relations).
10. **Fact:** Nationwide strike authorization by Costco workers.
– **Check:** Consistent with the summary.
– **Importance:** Lesser (relates to broader labor context but indirectly to docks).
– **Relevance:** False (separate industrial dispute).
These assessments focus on dock workers in the US, as specified in the question, and the likelihood of a strike based on the given context rather than unrelated international or industry developments.
Fact-checked summary:
The dock workers have agreed to salary increases, which is a critical part of the basis for returning to the bargaining table. On January 15, 2025, the dock workers’ union and the port resumed negotiations to discuss automation and other issues, an important detail that directly impacts the likelihood of a strike. Recently, a large-scale strike involving 45,000 International Longshoremen’s Association workers occurred on October 1, 2024, highlighting recent tensions in the sector, and the strike ended with a deal on October 3, 2024, demonstrating the capacity for quick resolution. Concerns about the potential impacts on the supply chain, inflation, and goods shortages add important economic context, potentially influencing negotiation pressures and affecting the likelihood of a strike in Q1 2025.
OpenAI gives a base rate of 0.2. The following were considered in order to produce this base rate:
The likelihood of a strike is assessed by considering recent negotiation developments and the sector’s historical precedent. The recent agreement on salary increases suggests a positive negotiation trend, reducing strike likelihood. However, contentious issues like automation remain. Historically, strikes have occurred in this sector, but the resolution of the October 2024 strike within days indicates a capacity for rapid compromise. Since negotiations have resumed as of January 15, 2025, with a focus on specific unresolved issues, there is an opportunity for resolution without strike action in Q1 2025.
Some potential divergent considerations that might effect the base rate:
Unforeseen political or economic pressures, such as a sudden economic downturn or political intervention, could increase tensions, impacting the probability of a strike. Additionally, global events affecting the supply chain might elevate urgency and affect negotiations, potentially leading to a union or management overreaction. These factors could cause deviation from historical trends and negotiation patterns. Furthermore, internal union politics or changes in leadership could alter the approach to negotiations, potentially affecting outcomes.
The following chain of events are necessary for the question to resolve positively:
– Breakdown in negotiations between the dock workers’ union and the port authorities. Unlikely
– Failure to reach an agreement on key issues such as automation and work conditions. Possible
– Escalation of tensions to a level that justifies a strike action. Possible
Querying Claude (AI predicts: 0.35 – confidence: 6)
Querying Mistral (AI predicts: 0.35 – confidence: 6)
Querying OpenAI (AI predicts: 0.3 – confidence: 6)
Question Type: Binary
Median from LLMs: 0.35
Base rate: 0.2 (from OpenAI)
SD: 0.02
Confidence: 6
Conf Mode: Normal
Mellers: 0.29
Reverse Mellers: 0.4
Theory of Mind: 0.5 (What did the LLMs think other LLMs predicted?)
Beta Distribution: 0.001
Close Type: A (B = cautious # closer to 50%; A/C = closer to extremes)
# LLM responses: 3
Model value: 0.35
We queried Claude, Mistral, OpenAI. A summary of what the responding AIs considered is as follows:
The prediction of whether a strike will occur in Q1 2025 is influenced by several factors. Key positive indicators include the recent successful salary negotiations and the quick resolution of the October 2024 strike, both suggesting a capacity for effective negotiation and compromise. Negotiations resumed on January 15, 2025, providing ample time for dialogue on contentious issues, such as automation and work conditions, which remain potential stumbling blocks. Historically, disputes in this sector are resolved efficiently, yet economic pressures and supply chain concerns could heighten the risk of strike actions. Furthermore, internal union dynamics and potential changes in negotiation strategies or leadership could influence the course of negotiations, with union leaders possibly feeling pressured to adopt a more aggressive stance or utilize strike threats as leverage. External factors, like unforeseen economic downturns or global events affecting the supply chain, could also disrupt negotiations and lead to outcomes that deviate from historical trends..
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?
* less than 172: 45%
* between 172 and 205: 35%
* between 206 and 240: 15%
* more than 240: 5%
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:
Most Recent Developments:
– Conservative Leader Pierre Poilievre has pledged to make “massive” cuts to Canada’s foreign aid budget to boost Arctic security, drawing some comparisons to Trump-style policies. Conservative strategists maintain this approach aligns with their platform despite the comparisons.
– The Conservative Party had record-breaking fundraising in 2024, nearly doubling the combined totals of the Liberal Party and NDP.
– The NDP is preparing for a possible March 10 election call as Conservative polling leads have started to narrow.
Electoral Landscape:
– Young voters are showing increased support for Conservative candidates, particularly in areas like Newfoundland and Labrador, indicating a potential shift in traditional voting patterns.
– In Ontario, voters remain split on key issues like healthcare, with no clear party emerging as the preferred choice to address the doctor shortage crisis.
– Mark Carney’s potential Liberal leadership bid could impact election timing, as he’s indicated he might call an early election if he wins the party leadership.
The Conservative Party appears to be in a strong position heading into potential federal elections, with robust fundraising and polling numbers, though there are signs of tightening in some areas. The party’s policy positions, particularly on foreign aid and Arctic security, continue to generate both support and controversy.
Summarizing outside newsfeed:
“Recent developments in Canadian politics suggest the Conservative Party is in a robust position, marked by record-breaking fundraising efforts in 2024 and a strong polling advantage. Conservative Leader Pierre Poilievre’s pledge to significantly reduce foreign aid to enhance Arctic security has drawn comparisons to Trump-style policies, sparking both support and controversy. Meanwhile, the NDP is bracing for a possible election call in March 2024 as the Conservative polling lead narrows. Shifts in voter demographics, particularly increased support from young voters in regions like Newfoundland and Labrador, indicate changing traditional voting patterns. In Ontario, voters remain divided on key issues such as healthcare. Additionally, Mark Carney’s potential bid for Liberal Party leadership could influence the timing of upcoming elections, as an early election could be called if he secures the leadership.”
Fact-checking news summary:
Based on the summary, here is a list of specific facts, their significance, accuracy, and relevance to the question of how many seats the Conservative Party will win in Canada’s next federal parliamentary election:
1. **Fact**: The Conservative Party is marked by record-breaking fundraising efforts in 2024.
– **True/False**: Unable to verify (context needed for confirmation).
– **Importance**: Important.
– **Relevance**: True. Fundraising is often correlated with the capacity to campaign effectively, which can influence election outcomes.
2. **Fact**: The Conservative Party has a strong polling advantage.
– **True/False**: Unable to verify (would need specific polling data for confirmation).
– **Importance**: Important.
– **Relevance**: True. Polling can be an indicator of potential election outcomes.
3. **Fact**: Pierre Poilievre pledges to significantly reduce foreign aid to enhance Arctic security, drawing comparisons to Trump-style policies.
– **True/False**: Unable to verify (requires crosschecking with Poilievre’s statements or Conservative Party platform).
– **Importance**: Lesser importance.
– **Relevance**: False. While relevant to understanding policy positions, this does not directly predict seat outcomes.
4. **Fact**: The NDP is bracing for a possible election call in March 2024.
– **True/False**: Unable to verify (context needed for confirmation).
– **Importance**: Important.
– **Relevance**: False. While the timing of an election is crucial, this fact doesn’t directly indicate seat counts for the Conservative Party.
5. **Fact**: Conservative polling lead is narrowing.
– **True/False**: Unable to verify (would need recent polling data).
– **Importance**: Important.
– **Relevance**: True. A narrowing poll gap could indicate tighter competition affecting seat predictions.
6. **Fact**: Increased support from young voters in regions like Newfoundland and Labrador indicates changing traditional voting patterns.
– **True/False**: Unable to verify (requires polling or electoral data).
– **Importance**: Lesser importance.
– **Relevance**: False. While demographic shifts can influence elections, this does not specify outcomes for Conservative seat gains.
7. **Fact**: Ontario voters remain divided on key issues such as healthcare.
– **True/False**: Likely true (subject to public opinion research).
– **Importance**: Lesser importance.
– **Relevance**: False. Division on issues does not directly forecast seat outcomes for the Conservative Party.
8. **Fact**: Mark Carney’s potential bid for Liberal Party leadership could influence the timing of upcoming elections.
– **True/False**: Unable to verify (would require internal Liberal Party information).
– **Importance**: Important.
– **Relevance**: False. While leadership changes can impact election dynamics, they do not directly indicate Conservative seat predictions.
9. **Fact**: An early election could be called if Mark Carney secures the Liberal leadership.
– **True/False**: Hypothetical and unable to verify without context.
– **Importance**: Important.
– **Relevance**: False. The timing of an election matters for strategy but does not directly determine Conservative seat counts.
Overall, the summary provides context and potential influencing factors, but specific forecasts for Conservative seat counts require precise polling and demographic analysis.
Fact-checked summary:
The fact-checking list does not include any items that are both definitely true and relevant in predicting the number of seats the Conservative Party will win in Canada’s next federal parliamentary election. Each fact requires further verification or does not directly indicate outcomes for the Conservative Party’s seat count. For accurate predictions, specific polling data and demographic analysis would be necessary.
Querying Claude (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: 4)
Querying Mistral (AI predicts: [“less than 172”: 0.5, “between 172 and 205”: 0.3, “between 206 and 240”: 0.15, “more than 240”: 0.05] – confidence: 5)
Querying OpenAI (AI predicts: [“less than 172”: 0.25, “between 172 and 205”: 0.35, “between 206 and 240”: 0.3, “more than 240”: 0.1] – confidence: 5)
Question Type: Multiple Choice
Confidence: 5
# LLM responses: 3
Model value:
* less than 172: 0.45
* between 172 and 205: 0.35
* between 206 and 240: 0.15
* more than 240: 0.05
We queried Claude, Mistral, OpenAI. A summary of what the responding AIs considered is as follows:
The AIs’ reasoning on the prediction of Canadian federal elections revolves around historical patterns of election results, current polling trends, and the prevailing political climate. They highlight the historical range of seats typically won by the Conservative Party, noting that significant shifts in seat distribution are uncommon. Current polling suggests a modest increase in support for the Conservatives, but not enough for a dramatic change. The multi-party nature of Canadian politics and stable political conditions contribute to this expectation. However, the prediction could be skewed by unforeseen events, such as major scandals, economic changes, or shifts in regional voting patterns, particularly in key areas like Ontario and Quebec. The lack of specific or recent polling data adds uncertainty, and potential future swings in public opinion or changes in party strategies could alter the anticipated outcomes.
Runtime for both questions: 113 seconds.
Past Multi-AI Oracle Forecasts
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).
