AI Forecast Tracker
← Back to scoreboard
FinanceBAC

Bank of America

Q2 2026 (July 14) — revenue $31.6B (+15% YoY on a restated base), net interest income $16.0B (+9%), net income $9.1B (+27%), EPS $1.21 (+34%); the efficiency ratio improved about 360bps to 59% on 6.6% operating leverage, with headcount at 211,304. Erica reached 24.6M active users (+23%) and has passed 3B lifetime interactions, and digital channels now account for 70% of sales across 50M active digital users. Management says AI will help hold headcount down but does not break out AI savings. In capital markets BofA came late to AI financing — a $520M OpenAI credit line in July and a dedicated senior AI-ecosystem banker team in September — and Bloomberg reported on September 21 that it lags rivals in AI debt underwriting.

AI Impact Score
6.9/10
↑ Positive
Scoring Breakdown
Sector Base
7
AI Revenue Exposure
5
Moat Durability
8
Disruption Risk (lower=better)
3
AI Adoption Maturity
8

Scenarios

Bull Case

Erica at 24.6M active users and 3B+ lifetime interactions is the largest deployed banking assistant in the US, and a $4B annual new-technology budget (inside roughly $13B of total tech spend) funds an AI-native consumer layer on top of the deepest deposit franchise in the country. Operating leverage of 6.6% in Q2 2026 shows the efficiency case turning into reported numbers.

Bear Case

Net interest income is still the largest revenue driver, and every 25bps of Fed cuts costs roughly $900M-$1.2B a year, so rate cuts can offset the efficiency AI is meant to deliver even though NII grew 9% in both Q1 and Q2 2026. Erica reduces call-centre cost without creating a revenue line, and no analyst has modelled incremental revenue attributable to it. In AI capital markets BofA is a late entrant — Bloomberg reported on September 21 that it lags rivals in AI debt deals. The 3,900-branch network costs roughly $7-8B a year to run, and AI-driven branch closures carry community-reinvestment and attrition costs that cap how fast efficiency can be captured.

Key Factors to Watch

  • ●Erica — 24.6M active users (Q2 2026, +23% YoY), 3B+ lifetime interactions
  • ●90%+ internal workforce AI adoption
  • ●$13B total tech spend; $4B earmarked for new strategic technology
  • ●Late entrant to AI capital markets — $520M OpenAI credit line (July 2026) and a senior AI-ecosystem banker team (September 2026), but lagging rivals in AI debt underwriting per Bloomberg

Score History

DateScoreDirectionNote
2026-09-266.9PositiveW39 rotation — HOLD 6.9. No post-August-8 AI adoption shift; summary moved to Q2 2026 (revenue $31.6B, Erica 24.6M active users) and three conflicting tech-budget figures reconciled. September AI-ecosystem banker push noted alongside Bloomberg's report that BofA lags rivals in AI debt deals.
2026-04-106.9PositiveScore 6.8→6.9 (rounding drift correction — dimensions unchanged, formula recomputed to match stored value)
2026-03-086.8PositiveScore 7.0→6.8 (formula reweight: sb 0.25→0.15, are 0.20→0.25, md 0.20→0.25, dr 0.20→0.25, aam 0.15→0.10)
2026-03-087.0PositiveScore 7.3->7.0 (are 7->5, dr 4->3, aam 9->8). External research cross-ref: Erica 50M users but AI revenue indirect (are=5), regulation-protected (dr=3), strong but only 15 GenAI use cases live (aam=8)
2026-03-087.3PositiveInitial assessment from batch 2 research

Finance Peers

Last researched: 2026-09-26

This is research and analysis, not financial advice. Scores reflect AI impact potential, not investment recommendations.