Introduction
Welcome to TheDayAfterAI's weekly AI Chatbot Stock Prediction Showdown, where we put six of the world's most widely used AI chatbots to the test on the same question: what will a chosen stock do over the next five trading days?
This week's subject is Kymera Therapeutics, Inc. (NASDAQ: KYMR), a clinical-stage biotechnology company developing targeted protein degradation therapies. KYMR enters the forecast window riding a wave of excitement from its late-breaking Phase 1b KT-621 data presented at the American Academy of Dermatology (AAD) 2026 Annual Meeting, but doing so against a backdrop of elevated macro volatility, a holiday-shortened trading week (U.S. markets are closed on Good Friday, 3 April), and significant short interest in the stock.
The six chatbots participating in this week's showdown are ChatGPT, Claude, Copilot, Gemini, Grok, and Perplexity. Each was given the same prompt asking for a prediction on KYMR's opening price on 31 March, closing price on 7 April, expected trading range, and the probability of a net price increase or decrease over the period. Their responses were collected pre-market on 31 March 2026.
As always, this article is published for informational and entertainment purposes only and does not constitute financial advice. Markets are unpredictable, and AI-generated forecasts carry the same fundamental uncertainty as any other analytical approach. Readers should conduct their own due diligence before making investment decisions.
Head-to-Head Comparison
The table below summarises each chatbot's headline predictions. Note that each chatbot accessed its own data sources at the time of the query, which can explain significant differences in their starting assumptions.
| Chatbot | Pred. Open | Pred. Close | Range Low | Range High | Up % | Down % |
|---|---|---|---|---|---|---|
| ChatGPT | $79.70 | $82.60 | $77.00 | $84.50 | 60% | 40% |
| Claude | $72.00 | $71.50 | $67.50 | $76.00 | 42% | 58% |
| Copilot | $80.00 | $84.50 | $72.70 | $87.30 | 62% | 38% |
| Gemini | $78.16 | $82.35 | $73.50 | $85.00 | 60% | 40% |
| Grok | $81.20 | $83.50 | $78.00 | $86.00 | 65% | 35% |
| Perplexity | $81.50 | $82.50 | $76.00 | $86.00 | 60% | 35% |
* Perplexity allocated a separate 5% probability to a flat outcome (±0.5% of open).
Consensus at a Glance
Five of the six chatbots lean bullish, predicting a net price increase over the five-day window with probabilities ranging from 60% to 65%. The consensus predicted opening price clusters around $79 to $81, and the consensus closing price sits roughly in the $82 to $84 range, implying a modest weekly gain of around 2% to 6%.
The clear outlier is Claude, which takes a dramatically bearish stance. Claude predicts an opening price of $72.00 and a closing of $71.50, assigning only a 42% probability of a price increase. Its analysis centres on a fundamentally different reading of the AAD conference: rather than focusing on KYMR's own positive KT-621 Phase 1b data, Claude highlights competitor presentations (Connect Biopharma's rademikibart, Aclaris's ATI-2138, Sanofi's amlitelimab) that it argues have reframed the competitive landscape against KT-621. Claude also emphasises the hostile macro backdrop, including the U.S.-Iran conflict, oil above $100/barrel, and a VIX above 30, as well as a pattern of heavy insider selling totalling $40.5 million over three months.
This divergence neatly illustrates a recurring theme in our showdown series: AI chatbots do not merely differ in their numerical estimates; they can arrive at fundamentally different narratives about the same stock, depending on which data points they weight most heavily and which sources they access in real time.
Key Drivers Identified Across All Models
1. KT-621 Phase 1b BroADen Data (AAD 2026)
All six chatbots identified the late-breaking AAD presentation as the single most important near-term catalyst. KYMR's KT-621, an oral STAT6 degrader targeting atopic dermatitis, showed 98% median STAT6 degradation in blood, 94% in skin, and approximately 63% mean EASI reduction over 28 days in Phase 1b results. The bullish majority views this as validation of the "oral dupilumab" thesis that has powered KYMR's 270% rally from its 52-week low. Claude, however, argues the data looked "early-stage and modest" next to competitor results presented at the same conference, and that markets price narratives over caveats about cross-trial comparisons.
2. Short Interest and Squeeze Mechanics
Every chatbot flagged elevated short interest, with estimates ranging from 12.6% to 18% of float and days-to-cover of 10 to 16. Gemini provided particularly granular FINRA data showing short volume ratios exceeding 80% on multiple recent days. The bullish models view this as fuel for a potential short-covering rally triggered by the positive clinical data. Claude acknowledges the squeeze potential but interprets the heavy short positioning as reflecting "deep bearish conviction."
3. Macroeconomic Headwinds
All models note the elevated VIX (around 28 to 31), geopolitical risk from the U.S.-Iran conflict, and oil prices above $100/barrel. However, there is wide disagreement on how much weight these factors carry for a biotech stock with company-specific catalysts. Grok and Copilot treat macro as a secondary consideration ("lighter weight"), while Claude and Gemini position it as a primary headwind that mechanically compresses biotech valuations through rising discount rates and systematic de-risking protocols.
4. Holiday-Shortened Week and NFP Gap Risk
All chatbots flag the Good Friday market closure and the unusual release of Non-Farm Payrolls on a day when equity markets are shut. This creates what Claude calls "extraordinary gap risk" for the Monday 6 April reopening. ChatGPT notes that weekly options expiration is pulled forward to Thursday 2 April, concentrating gamma effects and potentially increasing pinning or noise around nearby strikes.
5. Balance Sheet Strength
All models agree that KYMR's $1.6 billion cash position (runway into 2029) eliminates near-term dilution risk and provides a valuation floor. Even the bearish Claude model acknowledges that this "rock-solid balance sheet floor limits catastrophic downside."
Individual Chatbot Summaries
ChatGPT
Verdict: Modestly bullish, but volatile. ChatGPT predicts an open near $79.70 and a close at $82.60, representing roughly a 3.6% gain. It provides a detailed day-by-day table with open, high, low, and close estimates for each session. Its analysis is methodical, anchoring on the pre-market quote, layering in technical levels (support at $77.72/$75.95, resistance at $80.99/$82.50/$84.27), and flagging the NFP-on-a-holiday gap risk. ChatGPT gives the widest commentary on how the April 3 payrolls release could manifest as a Monday gap, and it notes weekly options expiration effects being pulled forward to Thursday.
Distinctive angle: The most structured day-by-day forecasting format among the six, with explicit invalidation levels ($77.7 as critical support, $81.0 as breakout confirmation).
Claude
Verdict: Bearish, with a fundamentally different thesis. Claude is the sole bear in the group, predicting a $72.00 open and a $71.50 close, with only a 42% probability of an increase. Its central argument is that KYMR's pre-market gap-down to $71.98 reflects a company-specific rerating driven by competitor data at the same AAD conference. It flags Connect Biopharma's rademikibart (96.6% EASI-75 at Week 52), Aclaris's ATI-2138 (77% EASI improvement at Week 12), and Sanofi's amlitelimab as having materially changed the competitive landscape. Claude also highlights $40.5 million in insider selling over three months, the U.S.-Iran diplomatic deadline coinciding with the market's reopening on 6 April, and a technical breakdown below all major moving averages.
Distinctive angle: The only chatbot to interpret the AAD conference as net negative for KYMR, and the only one to incorporate detailed competitor clinical data into the thesis. Also uniquely cites the Wolfe Research $60 bear-case target and provides enterprise value per share analysis.
Copilot
Verdict: Bullish with the highest closing price target. Copilot predicts an $80.00 open and an $84.50 close, the highest end-of-period target among the six, implying a 5.6% gain. It provides a clean day-by-day forecast with intraday ranges and assigns a 62% probability of upside. Copilot's rationale leans on the AAD KT-621 data as a fresh bullish catalyst, short-term momentum above key moving averages, and elevated implied volatility creating a wide but upward-biased range. Its approach uses options-derived expected moves (±7–9%) to frame the volatility envelope.
Distinctive angle: Most explicit use of options-implied volatility to derive the weekly range. Also the most optimistic on the magnitude of the post-AAD rally.
Gemini
Verdict: Bullish on fundamentals, cautious on macro. Gemini predicts a $78.16 open and an $82.35 close, a 5.3% gain with a 60% probability of upside. Its report is by far the longest and most detailed, running through an exhaustive multivariable framework covering moving average architecture, StochRSI analysis (at absolute 0, indicating extreme oversold conditions), FINRA short volume data, options open interest term structure, theta decay dynamics around the Good Friday holiday, M&A valuation floors, and macroeconomic cross-asset correlations. Gemini gives the most granular technical data, including specific short volume ratios (89.66% on 16 March) and moving average compression analysis.
Distinctive angle: The most technically comprehensive analysis. Uniquely highlights StochRSI at absolute zero, short volume ratios exceeding 80%, and the mechanical implications of quarter-end window dressing. Also the only model to discuss M&A as a structural valuation floor for mid-cap biotechs.
Grok
Verdict: Most bullish probability, moderate price target. Grok assigns the highest probability of upside at 65%, with a predicted open of $81.20 and close of $83.50, a 2.8% gain. Its analysis is structured around a clear factor-by-factor framework, weighting technicals and near-term catalysts most heavily. Grok notes bullish 5/10-day moving average crossovers, neutral RSI around 51, and a slight MACD buy signal. It provides detailed macro calendar awareness and observes that the UBS price target upgrade to $128 occurred before the AAD data, implying potential for further upgrades.
Distinctive angle: The most confident directional call (65% up). Provides the most concise, actionable summary format with clear factor weightings and a bottom-line paragraph.
Perplexity
Verdict: Cautiously bullish, acknowledging data limitations. Perplexity predicts an open near $81.50 and a close around $82.50, the smallest projected gain (approximately 1.2%) among the bullish models, with a 60% probability of upside. Uniquely among the six, Perplexity allocates a 5% probability to a flat outcome. Its analysis is notable for intellectual honesty about data limitations: it acknowledges that its accessible daily history only runs through mid-2025, making precise moving average and RSI calculations unreliable. Perplexity works around this by inferring the technical state from price action patterns and extrapolating a realised-volatility estimate. It provides detailed short interest analysis (9–10 million shares, 12–18% of float, 13–16 days to cover) and a balanced reading of insider selling patterns.
Distinctive angle: The most transparent about methodology limitations. Only model to assign an explicit flat-outcome probability. Most balanced treatment of insider selling (characterising it as "orderly profit-taking" rather than a red flag).
Why Do the Chatbots Disagree So Much?
The spread between the most bullish prediction (Copilot at $84.50 close) and the most bearish (Claude at $71.50 close) is a remarkable $13.00, or roughly 16% of the stock price. This is not a trivial difference. Three factors explain most of this divergence.
Different pre-market data. The chatbots accessed different data snapshots at query time. Five of the six observed a pre-market price in the $78 to $81 range, reflecting the post-AAD enthusiasm. Claude, however, observed a $71.98 pre-market figure, suggesting it accessed data reflecting a sharp gap-down driven by competitor news. This single data point difference of roughly $8 to $9 ripples through every subsequent calculation.
Different narrative framing. The bullish majority frames the AAD conference through the lens of KYMR's own positive data. Claude frames it through the lens of what the conference revealed about KYMR's competitors. Both framings are legitimate analytical approaches, but they produce directionally opposite conclusions.
Different macro weighting. While all chatbots acknowledge elevated macro risk, they differ substantially in how much weight they assign it. Grok and Copilot treat it as secondary to company-specific catalysts. Claude and Gemini treat it as a primary force that can overwhelm even strong idiosyncratic tailwinds, with Claude going furthest by incorporating specific geopolitical scenarios (U.S.-Iran diplomatic deadline on 6 April) into its price path.
What to Watch This Week
- Monday 31 March: The opening print will immediately tell us which chatbot's pre-market reading was correct. A gap-down toward $72 validates Claude's thesis; a stable open above $79 validates the bullish majority. Also watch Consumer Confidence and quarter-end rebalancing flows.
- Tuesday 1 April to Wednesday 2 April: ADP Employment, ISM Manufacturing, and jobless claims will set the macro tone. Watch the ISM "prices paid" component for real-time inflation signals from the oil shock.
- Thursday 2 April (last session before the break): De-risking and options positioning ahead of the Good Friday closure. Expect elevated volatility and reduced risk appetite in the final hour.
- Friday 3 April (markets closed): Non-Farm Payrolls release at 8:30 AM ET. All chatbots flag this as a major gap risk for the Monday reopening.
- Monday 6 April: The gap open after digesting NFP will be the single most volatile session of the week. Claude additionally flags the Trump-Iran diplomatic deadline on this date.
- Tuesday 7 April: The final session of the forecast window. Stabilisation is expected unless Monday produced an outsized move.
Methodology
Each chatbot was given an identical prompt requesting a five-trading-day price prediction for KYMR from 31 March to 7 April 2026. The prompt asked for a predicted opening price, predicted closing price, estimated trading range, probability of a net price increase versus decrease, and a rationale. No follow-up prompts were used. The chatbots accessed their respective real-time data sources at the time of query (pre-market on 31 March 2026). Responses were captured in full and are presented with minimal editing for formatting consistency.
The chatbots used in this edition are: ChatGPT (OpenAI), Claude (Anthropic), Copilot (Microsoft), Gemini (Google), Grok (xAI), and Perplexity. Readers should note that Gemini Deep Research mode has consistently failed to deliver within the pre-market window in previous editions of this series; this week's Gemini response was successfully obtained within the time frame.
