What you’ll learn in this article…
- Purdue's 2026 research says AI emphasizes people skills instead of replacing them.
- PR specialists earned a median $74,750 annually in May 2025.
- AI automates drafting but leaves persuasion, crisis judgment, and trust to humans.
Generative AI can now draft a press release, video script, or social post in seconds, leaving many communication majors to ask what remains for a human graduate in communication degree jobs. An October 2026 Purdue University article, “Are communication skills still important in the age of AI? Research says yes,” argues the answer is plenty.
That confidence needs a skeptical read. The trillion-dollar crisis cost and word-of-mouth dominance claims are industry assertions, not clean data. But the durable insight is that employers increasingly reward judgment: a human communicator with business communication skills decides whether an AI-generated message should be sent, and what it should say.
What the Purdue Article Argues About Communication Skills and AI
The conversation has shifted from "Will AI replace communication majors?" to a more useful question: which communication skills employers want become more valuable when generative tools handle the drafting. Purdue University's October 2026 article, "Are communication skills still important in the age of AI? Research says yes," argues that AI is expanding demand for communication expertise rather than shrinking it. The article points to McKinsey Global Institute research suggesting generative AI will likely emphasize people skills, and it cites Harvard Business Review's description of communication as a "nested" skill that many technical abilities depend on. Matt Charles, online graduate program director at Purdue's Brian Lamb School of Communication, frames human communication as the "constant, and much needed, companion" of disruptive technologies, comparing AI to the printing press and the internet.
The piece identifies Generative Engine Optimization, or GEO, as an emerging communication role focused on making content appear in AI-generated search results, one of the newer specialties in AI in Mass Media. That is a useful signal for graduate applicants, but readers should treat the source as an informed advocate: the university markets online communication programs, so its optimism about the field also serves its own enrollment goals.
Fact-Checking the Trillion-Dollar Crisis and Word-Of-Mouth Claims
The Purdue article repeats two big claims: a crisis communication mistake can cost millions per incident and up to a trillion dollars a year, and word-of-mouth is still the most effective way to sell. Both deserve a closer look.
The trillion-dollar crisis figure
No named original study, year, dataset, or calculation supports the trillion-dollar annual total. It is not clear whether the figure covers lost sales, market cap, litigation, remediation, or reputational damage. Company-specific crisis studies and brand-valuation estimates exist, but they do not add up to a worldwide annual total without a transparent denominator and double-counting controls. Verdict: unverified.
The word-of-mouth claim
Nielsen's 2021 Trust in Advertising Study1 surveyed more than 40,000 consumers and found 88% of global respondents trusted recommendations from people they knew more than any other ad channel; a U.S. version reported 89%.2 That measures stated trust, not actual sales effectiveness , a distinction central to building trust in communication. A separate Nielsen consumer analysis found only 30% of new brand triers used recommendations, while 45% found advertising useful or influential.3 McKinsey's 2010 word-of-mouth equity model found marketing-induced consumer-to-consumer word of mouth drove twice the sales of paid advertising in some categories, but the supplied source lacks sample size, markets, and econometric detail.4 Verdict: partly supported.
A quick soft-statistic check
When a figure circulates without a named methodology, original date, or primary dataset, treat it as an industry assertion. Circular citations are a red flag, the same signal used to spot fake news. Use that test before quoting crisis costs or word-of-mouth dominance.
Which Communication Tasks AI Automates, Augments, or Leaves to Humans
This table reflects our editorial judgment about how AI shifts communication work, not measured automation rates from a specific dataset. Real tools vary by employer, and many tasks move between columns as models improve. Use it as a conversation starter for identifying where your own expertise adds the most durable value.
| Subfield | Largely Automated | Augmented (Human Reviews) | Stays Human-Led |
|---|---|---|---|
| Public relations | First-draft press releases, media monitoring summaries | Crisis response drafts, sentiment analysis reports | Crisis judgment, reporter relationships, reputation repair |
| Journalism | Transcription, routine earnings recaps, sports box scores | Story discovery from datasets, summarizing public records | Source relationships, investigative judgment, fact-checking sensitive claims |
| Marketing | Subject line variations, ad copy batches, image tagging | Campaign performance narratives, SEO content outlines | Brand voice, word-of-mouth strategy, persuasive storytelling |
| Corporate communication | Internal FAQ drafts, meeting summaries, policy announcement templates | Employee messaging around change, executive talking points | Stakeholder trust, crisis decision-making, ethical framing |
| Technical writing | API documentation drafts, standard operating procedure formatting | User guide clarity checks, compliance terminology alignment | Interpreting user needs, simplifying complex systems, accessibility judgment |
| Interpersonal and intercultural communication | Translation drafts, scheduling and tone suggestions | Cross-cultural phrasing reviews, conflict de-escalation scripts | Trust building, nonverbal nuance, equity and power dynamics |
The Skills That Hold up: Persuasion, Crisis Judgment, and Trust
Can AI-generated messages persuade, reassure, and build trust as well as a trained human communicator? The 2026 evidence is mixed, but it clarifies which skills still hold.
Where AI is strong
Political persuasion research found chatbots and humans both beat placebo messages at five weeks, with a human edge near zero: 0.001 percentage points on binary outcomes and 0.059 scale points on continuous outcomes.1 One preregistered 2025 study set showed LLM messages shifted attitudes on polarized and less polarized policy issues.5 For perceived empathy, AI often scores higher: in six direct physician comparisons, LLMs were rated more empathetic in five (odds ratio 9.8),2 and one 556-person study found AI supportive messages rated more compassionate than expert human counselors even when labeled AI.3
Where judgment matters
But crisis research shows AI messages are timely and consistent while human messages are more sincere and credible, a gap that crisis communication experts are trained to manage. Sincerity carried the strongest trust link at 0.36;4 human-written disaster and risk alerts were also more credible than AI-led alerts.5 When health misinformation corrections were labeled "AI-written," persuasiveness dropped for both AI and human messages, especially when credibility was at stake.
These gaps matter in high-stakes fields. Political persuasion risks automated manipulation at scale, especially in social media and democracy. Health misinformation depends on ethos, not just pathos or logos. Intercultural nuance can be exploited when AI adapts identity cues to raise persuasion. Environmental messaging needs human accountability behind evacuation alerts.
AI handles volume and fluency. Reading a room, owning a decision, and staying accountable when a message lands wrong are the scarce skills.
Related Articles
Employer Demand and Entry-Level Pressure for Communication Graduates
The communication job market is splitting into two experience bands: routine drafting roles and judgment-heavy advisory roles. Federal projections mirror that split. Public relations and marketing positions are generally expected to hold or grow through 2034, while writing, editing, and some technical writing categories face slower or declining growth as automation absorbs first-draft work.
What employer demand shows
The World Economic Forum's 2025 Future of Jobs report, based on surveys of more than 1,000 companies, projects nearly 40% of job skills will change by 2030.1 Over three in five employers cite skills gaps as the biggest barrier to transformation.2 Analytical thinking, creative thinking, resilience, collaboration, and social influence stay in demand as soft skills for jobs, even as AI, big data, and technological literacy rank among the fastest-growing skill areas.1
McKinsey Global Institute places communication among eight high-prevalence skills across industries.3 Demand for AI fluency, using and managing AI tools, grew roughly sevenfold in the two years through mid-2025.3 That demand layers onto communication skills rather than replacing them.
The entry-level squeeze is real
Routine drafting, summarizing, formatting, and first-draft writing are the tasks most exposed to automation. Entry-level PR jobs and similar roles built around those outputs face the tightest pressure. What remains harder to automate is deciding what should be sent, adjusting the message for a specific audience, spotting reputational risk, and building trust.
Will AI replace communication jobs?
Most evidence points to reshaping, not replacing. WEF data shows about half of employers plan to reorient their businesses in response to AI, and 77% plan to upskill workers. Still, 41% expect to reduce workforces where AI automates tasks.1 For communication graduates, the stronger positioning is to pair core communication strengths with AI fluency and evaluation skills, not to compete with AI on volume drafting.
What Communication Professionals Earn Now
As of May 2025, the U.S. Bureau of Labor Statistics reports a national median annual wage of $74,750 for public relations specialists.
Their constant, and much needed, companion is human communication skills to articulate how best to utilize them and their outputs.










