NewMCP serverRead the guide

Diff the risk factors between two annual reports

Compares Item 1A across a company's two most recent 10-K filings and reports what was added, removed, expanded and cut, with word counts, separating substantive changes from boilerplate rewording.

Prompt

Copy and paste into your assistant
Summarize the differences between the risk factors from Lam Research's two most recent 10-K filings.

Output

Comparing FY2026 10-K (filed Aug 7, 2026, FYE Jun 28) against FY2025 10-K (filed Aug 11, 2025, FYE Jun 29). Item 1A grew from about 12,050 to about 14,450 words (+20%), same four categories, with one risk factor added and one removed.

Structural changes

  • Added: "We Use Artificial Intelligence in Our Business, and Challenges with Properly Managing Its Use Could Result in Reputational Harm, Competitive Harm, and Legal Liability". A new standalone factor covering internal AI and agentic AI deployment, competitors adopting AI faster, unsettled IP protection for AI-generated inventions, and employees leaking proprietary data into public AI tools.
  • Removed: the standalone "Epidemics, Pandemics or Outbreaks of Diseases" factor. Pandemic language survives only as a clause inside the facility-disruption risk.

Where the real expansion happened

Cybersecurity (+792 words, roughly doubled). The biggest single change. FY2025 described conventional threats plus a two-sentence nod to AI. FY2026 adds prompt injection, hallucinations, agentic AI learning to circumvent security controls, and LLMs retrieving data using a user’s credentials as a data-leakage vector. It also concedes that AI may discover and exploit vulnerabilities faster than Lam or its providers can patch them, and, new, that they may be unable to anticipate, identify or remediate incidents in time. Threat actors are now named explicitly: ransomware, state-sponsored intrusions, corporate espionage, credential harvesting.

International operations (+474 words). Three concrete new items, notable because they are framed as realized rather than hypothetical:

  • A Middle East conflict paragraph covering the conflict that began February 2026 among the US, Israel, Gulf states and Iran, including Strait of Hormuz restrictions and Red Sea vessel attacks, which "has adversely affected" supply and prices of oil, LNG, aluminum, helium, bromine and sulfur.
  • Chinese rare-earth export controls, including licensing on non-Chinese-made products containing Chinese-origin rare earths, partially suspended until November 2026.
  • Steel and aluminum tariffs that "increased our manufacturing costs in fiscal year 2026, which adversely impacted our margins."

Demand cyclicality (+259 and +154 words across two factors). AI is now named as the explicit driver of customer capex in both memory and non-memory, then reframed as a downside risk, with a new list of reversal scenarios: slower AI adoption, gains in compute efficiency, oversupply of AI infrastructure, and government regulation of AI. A mirror-image risk was added on the upside: failure to scale manufacturing and hiring fast enough to meet AI-driven demand costs them market share.

Legal exposure. The IP claims factor was retitled to add "Misuse of Third-Party Information" and now warns that law enforcement may investigate or charge Lam or its employees over improper receipt or misuse of designated technology or third-party confidential information under national and economic-security measures, with civil or criminal liability. The regulatory factor adds a candid admission: "there have been, and may continue to be, instances of our compliance policies and procedures not being effective."

Competition (+108 words). New scenario where competitors acquire or take control of suppliers of key IP, materials or components, cutting off Lam’s access.

Removed or softened

  • The entire BEPS 2.0 global minimum tax discussion was deleted from the tax risk factor (−119 words).
  • The restructuring and business-transformation factor was trimmed by about 20%.

Cosmetic but pervasive: "operating results" was globally rewritten to "results of operations", often with "and financial condition" or "and cash flows", which changed five headings without changing meaning. Revenue disclosures updated: China at 34%, 34% and 42% for FY26, FY25 and FY24, non-US at about 93% in each year.

Sources: LRCX FY2026 10-K·LRCX FY2025 10-K

When to use this prompt

  • Reading a new annual report without re-reading the whole of Item 1A.
  • Finding the risks a company started treating as realized rather than hypothetical, which is where new disclosure usually appears first.
  • Tracking how an industry-wide theme, such as AI or export controls, enters formal disclosure across a sector.

What you get back

  • Added and removed risk factors named in full, so a deletion is as visible as an addition.
  • Word-count deltas per factor, which rank the changes by how much the company actually rewrote.
  • Substantive changes separated from boilerplate. Here, a global rewrite of "operating results" to "results of operations" changed five headings and meant nothing.
  • Direct quotes where the wording matters, such as a risk framed in the past tense rather than the conditional.

Variations

  • Change the company, or run the same diff across a peer group to find a shared new risk.
  • Compare across five years instead of two to trace when a theme first appeared.
  • Diff a different item, such as Item 7 MD&A or Item 3 Legal Proceedings.
  • Ask only for factors that changed from conditional to past tense, which flags realized risk.

Filings used

  • Lam Research FY2026 Form 10-K, filed August 7, 2026, fiscal year ended June 28, 2026.
  • Lam Research FY2025 Form 10-K, filed August 11, 2025, fiscal year ended June 29, 2025.