In the question below, a paragraph is split into several parts. Examine each part for grammatical accuracy and select the option that correctly identifies which parts are grammatically correct or incorrect.


(A) Artificial intelligence is transforming many industries over the past decade. (B) If organizations invest in transparent data practices, users will trust AI systems more. When models are trained on diverse datasets, they tend to generalize better. (C) Had the team validated the model earlier, it could had prevented the bias incident. (D) As regulations tighten, companies (E) will need to document how their algorithms make decisions.

Correct option is B

Explanation & Corrections
· (A) — Incorrect (Tense consistency)
· Error: is transforming … over the past decade
· Rule: With time frames like over the past decade, use present perfect (continuous) for an action that started in the past and continues now.
· Correction: “ Artificial intelligence has been transforming many industries over the past decade.”
· (“has transformed” is also acceptable.)
· (B) — Correct (First conditional)
· Form: If + present simple, will + base verb.
· “If organizations invest…, users will trust…”
· (C) — Incorrect (Third conditional auxiliary)
· Error: could had prevented
· Rule: Third conditional uses would/could/might have + past participle.
· Correction: “Had the team validated the model earlier, it could have prevented the bias incident.”
· (D) — Correct (Time clause + future)
· Present simple in the time clause ( As regulations tighten) + will in the main clause is standard.
· (E) — Correct
· Sentence is grammatically sound and idiomatic.
Corrected paragraph (A & C fixed):
(A) Artificial intelligence has been transforming many industries over the past decade. (B) If organizations invest in transparent data practices, users will trust AI systems more. (C) Had the team validated the model earlier, it could have prevented the bias incident. (D) As regulations tighten, companies will need to document how their algorithms make decisions. (E) When models are trained on diverse datasets, they tend to generalize better.