paint-brush
The Impact of Generative AI on Tech Outsourcing Modelsby@devinpartida

The Impact of Generative AI on Tech Outsourcing Models

by Devin PartidaAugust 30th, 2024
Read on Terminal Reader
Read this story w/o Javascript
tldt arrow

Too Long; Didn't Read

Research suggests executives could upset some users by replacing customer service representatives with AI chatbots. One concern is that developers train these products on human-generated data, but people make mistakes. The rise of generative AI has also suggested that tech outsourcing models will change by altering the duties that those hired assume.
featured image - The Impact of Generative AI on Tech Outsourcing Models
Devin Partida HackerNoon profile picture

Generative AI has upended the tech industry, and many wonder how much it will impact formerly outsourced roles. These questions remain largely unanswered since companies are in the relatively early stages of adoption. What’s known so far about this much-discussed topic, and how can people respond to the changes before it’s too late?

Using AI for Customer Support

Do generative AI chatbots threaten customer service roles? That question has sparked much debate, both from people who assist others with essential work and those who outsource such duties when needed. The answer is complex.


Firstly, research suggests executives could upset some users by replacing customer service representatives with AI chatbots. Consider a December 2023 study that found 64% of respondents would prefer companies not to use AI for customer service. Additionally, 53% of participants would consider switching to competitors if they heard of businesses beginning to use AI for that purpose.


Leaders who are considering replacing outsourced customer service representatives with generative AI tools must understand the possible tradeoffs, too. One concern is that developers train these products on human-generated data, but people make mistakes. Not surprisingly, it’s easy to find examples of generative AI tools providing seemingly correct information with inaccuracies mixed into the content.


That issue could be a recipe for disaster if a customer receives wrong information from a chatbot and trusts it. However, it is a difficult problem to solve due to the source material that shapes how generative AI tools work. When developers built the first version of ChatGPT, 22% of the training data came from Reddit. Although many users contribute genuinely helpful and valuable content, others post questionable, sarcastic, and misleading replies.

Changing Outsourced Responsibilities

Outsourcing has become a popular option that allows people at companies to focus on their core competencies. Although executives may initially resist giving some duties to others — believing no one can do the tasks better than themselves — they quickly realize that outsourcing makes good business sense because it gives them more time and energy to focus on the areas where they truly excel.


Those interested in outsourcing must determine which tasks third-party companies should handle. From there, they can look forward to numerous benefits, such as reduced costs and more flexibility. The rise of generative AI has also suggested that tech outsourcing models will change by altering the duties that those hired assume.


For example, since AI is in the relatively early stages of real-world use, many applications still require humans to work behind the scenes, labeling data, supervising systems, and otherwise ensuring tools perform as expected. Similarly, some content creators and web developers use generative AI to support their primary duties. Still, those individuals learn that they must carefully check the output and — ideally — only use it for inspiration rather than copying everything verbatim from AI tools and using it in their work.


Interestingly, a study also revealed that those handling outsourced tasks may outsource their responsibilities to generative AI tools. While conducting a June 2023 study, researchers calculated the likelihood of workers on the Amazon Mechanical Turk platform using large language models for text production tasks. The findings indicated up to 44% of users did so when completing their assignments.

Interesting Developments Ahead

It is too early to say for sure whether generative AI will ultimately replace specific commonly outsourced tasks. However, the more likely outcome is that those who accept these responsibilities must evolve with the changing times by pursuing skill-growth opportunities.


Similarly, the wisest workers who meet outsourcing needs will recognize the value of figuring out the most feasible ways to work with technology rather than resisting it. Then, they will be as prepared as possible to respond to possible job shakeups and take new positions to remain relevant in a fluctuating landscape.