6 AI Chatbots Predict Alcoa (AA) Stock After Iranian Strikes Remove 9% of Global Aluminium Capacity — Mar 30–Apr 6, 2026

Introduction

Welcome to another edition of the AI Chatbot Stock Prediction Showdown, TheDayAfterAI News’s recurring series where we pit six of the most widely used AI chatbots against each other in a head-to-head stock price forecasting challenge. Each chatbot receives an identical briefing and is asked to produce a comprehensive short-term price prediction for a single ticker over the coming five trading days. No human editing is applied to the forecasts themselves; the raw analytical output from each chatbot is preserved in full.

This week’s subject is Alcoa Corporation (NYSE: AA), one of the world’s premier aluminium producers, entering what may be the most consequential trading week of 2026 so far. The five trading sessions span Monday 30 March through Monday 6 April (with markets closed on Friday 3 April for Good Friday). The catalyst driving extraordinary interest in AA is a major geopolitical supply shock: Iranian strikes on two of the world’s largest aluminium smelters in the Gulf region over the weekend of 28–29 March, removing approximately 9% of global aluminium capacity at a time when LME inventories already sit at historic lows.

The six chatbots participating in this week’s showdown are ChatGPT, Claude, Copilot, Gemini, Grok and Perplexity. Each was queried before market open on Monday 30 March 2026 and asked to provide a predicted opening price, predicted closing price at the end of the period, an estimated trading range, and a probability assessment of whether the stock would finish higher or lower than where it opened.

Why Alcoa, Why Now

Alcoa’s selection for this week’s showdown was driven by a rare convergence of catalysts that stress-test every dimension of an AI chatbot’s analytical capability. The Iranian strikes on Emirates Global Aluminium (EGA) and Aluminium Bahrain (Alba) represent a direct, physical destruction of competitor production capacity. LME aluminium surged roughly 6% to approximately $3,492 per ton on 30 March, its highest level since the 2022 Russia-Ukraine spike. The US Midwest premium hit a record $1.075 per pound.

Simultaneously, the broader macro environment is hostile. The S&P 500 had fallen roughly 7% year to date, the VIX was elevated near 31, oil was approaching $100 per barrel, and the February nonfarm payrolls report (released on Good Friday with markets closed) had printed a shocking loss of 92,000 jobs. The economic calendar for the week was packed: Powell speaking Monday, JOLTS Tuesday, ISM Manufacturing and ADP Wednesday, jobless claims Thursday, and the NFP reaction deferred to Monday 6 April.

Adding a final layer of binary risk, President Trump’s temporary pause on military strikes against Iranian energy infrastructure was set to expire at 8:00 PM ET on Monday 6 April, making the final session of the forecast window an extraordinarily volatile one.

Headline Predictions at a Glance

Note: Copilot and Gemini midpoints are used where ranges were given. Perplexity was the only chatbot to assign a net bearish probability for the open-to-close move. All figures are as reported by each chatbot prior to market open on 30 March 2026.

Key Observations

The Pre-Market Gap Divide

The most striking divergence among the six chatbots is how they handled the pre-market data. ChatGPT, Claude and Perplexity all incorporated the 8–10% pre-market surge into their opening price estimates, projecting AA to open near $64. Grok landed in the middle at roughly $62.25. In sharp contrast, Copilot and Gemini appeared to largely ignore or underweight the pre-market move, projecting opening prices near the prior Friday close of $58.41.

This is a significant methodological difference. If the stock does open near $64 (as pre-market trading strongly suggests), Copilot and Gemini’s forecasts will be structurally mispriced from the first tick. The pre-market gap is not speculative; it reflects actual transactions driven by a verified geopolitical catalyst. The chatbots that failed to incorporate it were either working with stale data or applying an overly conservative anchor to the prior close.

Bullish vs Bearish: The Probability Spread

Five of the six chatbots lean bullish on the open-to-close trajectory, but with notably different conviction levels. Gemini is the most bullish at 65% probability of a price increase, followed by Copilot at 63% and Grok at 62%. ChatGPT and Claude cluster together at 55–56%. Perplexity stands alone as the sole bearish voice, assigning a 55% probability that AA will close below its Monday opening price by 6 April.

Perplexity’s contrarian stance is grounded in a well-reasoned argument about mean reversion after event-driven gaps. Historically, large overnight gaps driven by news catalysts do tend to experience partial retracement over the subsequent days, even when the underlying fundamental driver remains intact. This is a defensible analytical framework, though it does carry the risk of being overridden if the aluminium supply shock escalates further during the week.

Range Estimates and Volatility Assumptions

The predicted trading ranges reveal how each chatbot models volatility. Gemini projects the widest range at $55.00 to $64.50, a spread of $9.50 or roughly 16% of the opening price, reflecting its emphasis on the VIX environment and macro data risk. Grok similarly models a wide $57.00 to $66.00 band. ChatGPT and Claude project comparable ranges of approximately $6.30 and $6.50 respectively, centred higher given their elevated opening price assumptions.

Copilot’s range of $57.80 to $61.00 is the narrowest at just $3.20, which feels implausibly tight for a stock with a beta of 1.72–1.82 in a week featuring a major geopolitical catalyst, a heavy economic data calendar, and VIX above 30. This may reflect Copilot’s apparent failure to fully price in the pre-market move, resulting in a muted volatility assumption.

Analytical Depth Comparison

Individual Chatbot Profiles

ChatGPT

ChatGPT delivered one of the most operationally useful analyses, combining a clear session-by-session forecast with well-sourced reasoning at each step. Its narrative arc of “shock gap up, partial fade, midweek consolidation, Monday breakout attempt” is intuitive and well supported by its technical and macro analysis. ChatGPT was notably precise on the macro calendar, correctly identifying the Good Friday closure, the deferred NFP reaction, and the implications of the April 2 options expiry. It also flagged the $150 million revenue timing issue from Alcoa’s inventory repositioning, a detail that only Claude matched. The main limitation is a relatively narrow probability spread (56/44), which may understate the genuine uncertainty in a week this volatile.

Claude

Claude produced the most detailed fundamental analysis of the group, with particularly strong coverage of the smelter strike mechanics, the Section 232 tariff dynamics, and Alcoa’s data centre site monetisation strategy. Its day-by-day forecast closely mirrors ChatGPT’s, projecting an initial gap-up, midweek pullback, and a positive close on 6 April driven by the NFP reaction and Iran deadline convergence. Claude was the only chatbot to assign explicit probability scenarios to three geopolitical outcomes (escalation 30%, status quo 45%, peace 25%), which adds genuine analytical value. Its closing estimate of $65.00 is the second-highest of the group. One notable observation is Claude’s medium-low confidence self-rating, an unusual degree of epistemic humility that investors may actually find reassuring.

Copilot

Copilot’s analysis is competent but visibly handicapped by a failure to incorporate the pre-market data. Its predicted opening of $58.50 (essentially flat to the prior close) and closing of $59.65 would represent only a modest 2% gain for the week, dramatically underestimating the magnitude of the aluminium supply shock. The technical analysis references an “ascending triangle” and standard moving average levels, but these are rendered largely irrelevant by the overnight gap. Options and sentiment coverage is thin, and the macro calendar section, while present, lacks the granularity of the stronger entries. If the pre-market pricing holds, Copilot’s forecast will be the most disconnected from reality on day one.

Gemini

Gemini produced by far the longest and most academically rigorous analysis, running to approximately 5,000 words with 68 cited sources. Its coverage of the aluminium supply chain mechanics (potline freezing timelines, regional capacity percentages, smelter-by-smelter impact) is unmatched. The options microstructure section, detailing specific gamma walls, institutional put-selling at the $51–55 strikes, and call accumulation at $65/$75/$80, is the most sophisticated of any entry. However, Gemini’s opening price prediction of $58.20 is the lowest of the group and appears to ignore the pre-market surge entirely, anchoring instead to pre-market VWAP data from a very early window. This is a recurring pattern we have observed in the Showdown series: Gemini’s research depth is consistently best-in-class, but its headline price estimates sometimes lag real-time market conditions. Its closing estimate of $61.80 implies a 6.2% gain from its own opening estimate, the largest implied move of any chatbot, which partly compensates for the low starting point.

Grok

Grok occupies the middle ground in both price estimates and analytical depth. Its opening range of $61.50–$63.00 splits the difference between the pre-market-aware group and the pre-market-unaware group, suggesting it partially incorporated the gap but applied some discount. The closing range of $62.00–$64.00 at 62% bullish probability is reasonable. Grok’s analysis is well structured, with clean separation of technical, sentiment, fundamental, macro and risk categories. It is the most concise of the six entries, which has both advantages (readability, clear hierarchy of factors) and disadvantages (less supporting detail on options flow and macro calendar specifics).

Perplexity

Perplexity is the contrarian of the group and, depending on how the week plays out, may prove either the most insightful or the most wrong. Its 55% bearish probability is grounded in a statistically defensible argument: large news-driven gaps frequently experience partial mean reversion over a multi-day window, even when the underlying catalyst remains intact. Perplexity explicitly frames its estimates as “scenario weights” rather than precise forecasts, a degree of intellectual honesty that distinguishes it from the more confident-sounding entries. Its closing estimate of $61 (down from a $64 open) would still represent a meaningful gain from the prior Friday close. The practical trading framework at the end, focused on monitoring early intraday acceptance or rejection of the gap level, is genuinely useful for active traders.

The Consensus View

Averaging across all six chatbots (using midpoints where ranges are given), the consensus forecast for Alcoa over the five-session window is as follows:

The consensus is mildly bullish, reflecting the genuine strength of the aluminium supply shock partially offset by the hostile macro environment, elevated VIX, and the risk of a geopolitical de-escalation that would rapidly unwind the commodity premium. The tight implied move of +0.9% from open to close masks significant intra-period volatility; all chatbots expect daily swings of $2–3 or more.

Where the Chatbots Agree

Despite the divergences, several themes emerged consistently across all six analyses. All chatbots identified the aluminium supply shock as the dominant driver of AA’s price action for the week. All recognised that the Good Friday market closure creates an unusual structural dynamic, with the NFP reaction deferred to Monday 6 April. All flagged elevated VIX and broad equity market weakness as headwinds that could partially offset the commodity tailwind. All noted that Alcoa’s fundamentals are strong, with Q4 2025 earnings beating estimates by a wide margin and Q1 2026 earnings (due 16 April) likely to benefit from surging aluminium prices. And all acknowledged that the April 6 Iran deadline represents a significant binary risk event.

Where the Chatbots Disagree

The primary disagreement centres on whether the pre-market gap will hold through the week. ChatGPT, Claude and Perplexity assume the gap is real and price their forecasts accordingly; Copilot and Gemini effectively ignore it. Grok splits the difference. This is not a minor analytical quibble; it produces a $6 spread in opening price estimates, which cascades through every downstream calculation.

The secondary disagreement is directional. Five chatbots lean bullish; Perplexity leans bearish. Perplexity’s argument is not that the aluminium story is wrong but rather that large gaps tend to partially retrace, that the broader market is fragile, and that profit-taking pressure will dominate over a five-day window. This is a sophisticated, data-informed contrarian view that deserves respect even if it ultimately proves incorrect.

Editorial Notes

All six chatbots were queried on the morning of 30 March 2026, before US market open. Each received an identical prompt requesting a comprehensive five-trading-day forecast covering price action, technical analysis, options flow, sentiment, fundamentals and macro calendar impact. No follow-up prompts or clarifications were issued.

Gemini was queried using Deep Research mode. As has been noted in previous editions of this series, Gemini Deep Research consistently produces the most comprehensive and best-sourced output but requires significantly more processing time than the other chatbots. In time-sensitive applications such as pre-market analysis, this latency can be a practical constraint.