Generative AI is lowering the effort required to create workplace material, but employees report spending substantial time correcting what others send them. New research raises questions about how organizations account for the full cost of AI-assisted work.
An AI-assisted document can reach a colleague before its sender has read it in full. Its polished appearance offers little indication of whether the contents have been checked, yet the recipient may be expected to rely on it. As generative AI becomes embedded in routine workplace communication, the time saved creating material is becoming easier to identify than the effort required to make it usable.
Use.AI surveyed 8,316 adults who use AI across the United States, the United Kingdom, Canada, Australia, the European Union, and Latin America. 62% said AI had made it easier to produce work they previously would not have considered worth doing, including internal memos, presentations, and code.
Lower production costs allow employees to pursue tasks that once seemed too time-consuming to justify. Some may deliver genuine value. Others can enter a workplace's normal flow of communication before anyone has established whether they deserve the attention they will require.
The survey found that 34% of respondents had forwarded AI-assisted work without fully reading it themselves. This creates uncertainty at a familiar point in workplace communication: receiving a document ordinarily carries an expectation that its sender understands the material being shared. With AI-assisted output, that expectation can become unreliable.
Problems often emerge after the material arrives. 48% reported receiving AI-assisted work that appeared finished but required substantial correction, while 43% said they had been forced to redo AI-assisted work received from others.
For 39%, checking AI-assisted material had taken longer than completing the underlying task themselves would have taken. The finding illustrates a potential weakness in measuring AI productivity through drafting speed. A task can become faster for its original producer while consuming more attention from the person responsible for making the result usable.
The additional work can affect how quickly an organization operates. 41% of respondents reported spending more time reviewing, filtering or verifying content than before AI entered their workflow. Another 37% said problems involving AI-assisted material had delayed decisions.
A delayed decision extends the consequences beyond the document itself. Information awaiting correction can hold up work that depends on it, while colleagues spend time determining which parts of a finished-looking deliverable can be trusted. The effort may be dispersed across several people, making the cost difficult to attribute to the original task.
Oversight appears uneven. 58% of respondents reported that their workplaces lacked clear rules for checking AI-assisted material before it was shared. Without explicit expectations, employees may apply different standards to the same kind of output, leaving recipients uncertain about what has already been verified.
The survey measures separate self-reported experiences and does not establish that the increase in content production directly caused the reported rework or decision delays. It also does not calculate whether those costs outweigh the productivity gains AI provides. The findings nevertheless expose a practical limitation in evaluating efficiency through individual output alone.
A document generated in minutes may still require extensive human attention before it can inform a decision. Organizations can capture meaningful benefits from making previously impractical work economical, but those benefits depend on the quality of what ultimately reaches others.
The economic value of AI-assisted work depends on the effort required to make its output usable across the organization. Faster drafting is only one part of that calculation. As producing material becomes easier, the standards governing what employees circulate will help determine how much productive work AI actually saves.
About Use.AI:
Use.AI is a universal AI assistant that aggregates the world's leading large language models into one unified and seamless experience. It provides users with a single point of access to the most advanced AI capabilities available today, from complex problem-solving to creative content generation. By bridging the gap between multiple AI technologies, Use.AI empowers users to enhance their productivity and leverage cutting-edge intelligence in their daily workflows.
Use.AI is a universal AI assistant that aggregates the world's leading large language models into one unified and seamless experience. It provides users with a single point of access to the most advanced AI capabilities available today, from complex problem-solving to creative content generation. By bridging the gap between multiple AI technologies, Use.AI empowers users to enhance their productivity and leverage cutting-edge intelligence in their daily workflows.