In recent years, generative AI has seen an exponential rise in adoption across industries. Powered by advanced algorithms, vast datasets, and increased computing power, its creative potential is being leveraged for content generation, user personalization, and innovative research. This widespread adoption highlights generative AI’s transformative role and the growing trend of human-AI collaboration.
The market size of AI is expected to grow at a compound rate of 37.3% annually for the next seven years, contributing up to $15.7 trillion to the global economy by 2030. Generative AI—the kind that is creative and collaborative—constitutes a bulk of the size.
Generative AI has come a long way, from creating simple chats to generating lengthy, grammatically sound text, like the opening paragraph here—which is entirely and as it is generated with AI. It also has a long way to go. This juncture therefore is an apposite vantage point for us to consider the impacts, trends, and challenges associated with generative AI.
Impact of generative AI on the economy
Most business leaders (74%) consider generative AI as a top emerging technology that will impact their business in the next couple of years. Generative AI will take over most redundant and menial tasks and allow employees to focus on core and critical parts of work.
Technologies related to generative AI will increase productivity and economic output. According to LinkedIn’s “Future of Work” report, 47 percent of US executives believe that using generative AI will increase productivity. The same report also states that 84 percent of US workers could leverage generative AI to automate at least a quarter of repetitive tasks and enhance productivity. And generative AI is expected to add $7 trillion to the global GDP, according to Goldman Sachs.
Generative AI will have a significant impact on most industries—mostly positively. Industry experts suggest that, by 2030, around 30 percent of jobs could be automated thanks to generative AI. Jobs most exposed to this include office support and production work, where 87% and 82% of tasks, respectively, could be potentially automated.
Automation will not, however, render humans superfluous. Instead, it might lead to the transformation of job roles, increased efficiency, and creation of new opportunities. For instance, while high-end generative AI tools like Claude and Hypotenuse AI have created doubts about the profitability of hiring writers, AI-content editing has gained significant attention from relevant domains.
The impact of generative AI on the quality of jobs is of greater significance than the quantitative impacts. A survey by Upwork reveals that a majority of C-suite executives said their organizations will hire more professionals of all kinds because of generative AI—this is so that they can adapt to the technology and also deal with new tasks that will arise as a result. This should allay fears that generative AI will put humans out of work.
Protect your brand from the risks of AI-generated text
Get expert human editors to enhance AI-written content
Trends in generative AI adoption
With the numerous benefits it offers, generative AI continues to see widespread adoption. According to another survey by McKinsey, 79 percent of all respondents say that they have used generative AI either for work or outside of work, with 22 percent using it regularly for work. And 75 percent of the participants say that they expect generative AI to have a significant or disruptive impact on the nature of their industry in the next three years.
Marketing and sales is the function where generative AI is most widely used, with 14 percent of respondents saying their organization uses it for tasks associated with them. Product & service development and service operations are two other functions for which it is most widely used.
Generative AI penetration is the least for manufacturing, supply chain management, and HR. Only two percent of C-suite executives say their organization uses generative AI for manufacturing-related tasks.
Marketing and advertising, again, is the industry that is most keen to adopt generative AI. This is not surprising because marketing is a field that involves lots of tasks that generative AI excels at. Gartner estimates that about 30 percent of outbound marketing content will be generated using AI, by 2025.
Software engineering, though, is the field with the most share of skills augmentable by generative AI. According to the LinkedIn report mentioned previously, only three percent of software engineering tasks need specialized human skills.
Despite this, cross-functional adoption of generative AI remains measly. Less than a third of respondents in the survey conducted by McKinsey said that their organizations have used generative AI for more than one business function. For three functions or more, the share is only 16 percent. These figures have remained steady since 2021, indicating that there is much scope for expansion.
Adoption of generative AI results in a reduction in cost and increase in revenue. The impact is most pronounced in manufacturing, where 55 percent of respondents say their organizations saw a decrease in cost, and 66 percent, an increase in revenue, as a result of adopting generative AI.
Manufacturing is just one example that benefits from incorporating generative AI in its process. Enterprises could use generative AI to deliver operational efficiency, enhance product offerings, offer new experiences, or enforce operational changes.
Interest in generative AI has therefore soared and so has investment in AI-related technologies. Goldman Sachs projects that investment in AI could reach up to 4 percent of GDP in the US in the next 10 years, with the generative kind taking up a major share. A PwC survey shows that integrating new technologies into existing business models over the next two to five years is the chief aim of most executives. And nearly half of them plan to invest specifically in generative AI.
The increase in investment and the wider adoption of generative AI will lead to a fundamental shift in the nature of work and business operations. They will have a significant impact on research and development, product development, and business operations. The result of all this is higher innovation, increased efficiency, and higher customer satisfaction; and of course the flip side: concerns with inaccuracy, bias, data privacy, and losing human connection, among others.
Why are human-powered product descriptions better than AI copywriting?
Challenges, concerns, and prospects of generative AI
Generative AI has delivered much and still promises plenty. The challenges and concerns that it is fraught with are in nearly equal proportion. Some of these challenges are intractable and some are even insoluble.
Substantial quantities of data are required to train and assess the generative AI models. Gathering, data annotation, and refining raw data can be tedious, costly, and difficult. Moreover, training generative AI models is expensive and energy-intensive. Well-annotated training data, for instance, can help learn the model with more efficiency.
Generative AI models are only as good as, among others, the training data and computational resources—both of which are finite—and so they will always occasionally falter. Therefore, though organizations are keen to adopt generative AI, they also remain wary of it.
McKinsey’s “State of AI” report says that 56% of organizations consider inaccuracy a relevant risk with generative AI. Cybersecurity is the second major concern: 53% of respondents consider it a risk. Intellectual property rights violations and regulatory compliance are the next two.
Workforce displacement is another associated risk. 34% of organizations regard it as a risk with 13% working toward mitigating it. Generative AI may not necessarily render humans jobless but it necessitates them to reskill and re-orient their strategy. Human traits such as analytical judgment, flexibility, and emotional intelligence are essential to counterbalance generative AI.
There is a gendered aspect to this, too. Women are two times more likely to be affected by automation—since the types of work they generally do are also those most susceptible to AI—as a result of generative AI, according to the ILO. Organizations must not turn a blind eye to this. They need to ensure that efforts to mitigate job displacement by generative AI are undertaken in proportion to the degree of its impact on workers.
Want to make your AI system qualitatively impactful?
We make that possible with bespoke data labeling services
The necessity for humans in the loop of generative AI
Impressive as they are, generative AI is not completely failsafe, and not likely to be so anytime soon. It is constrained mostly by the availability of data. AI still requires human oversight and involvement to ensure its outputs are ethical, accurate, and aligned with intended goals.
The many ways generative AI models can go awry and their ramifications necessitate human involvement and oversight—not just during training (in the loop) but also verifying the output (outside the loop). Humans are needed to carefully curate the datasets used, prepare them for AI training through annotation, consider potential biases and gaps, set the objectives and constraints for generative models, test the system, flag inaccurate or inappropriate outputs, and provide additional feedback to enhance the system’s performance.
For instance, for high-stakes applications of AI, like digital revenue recovery software, ESG impact reporting platform, or smart parking applications, manual oversight by domain experts is essential to validate AI-generated insights and recommendations. Even for lower-risk applications like creating marketing copy or conversational bots, human reviewers can spot problematic outputs before public release.
Thus, a humans-in-the-loop approach is not just ideal when working with AI; it is pragmatic.
As generative AI becomes more prominent in our lives, particularly within the realm of technology blogs, ensuring that the technology is accurate, fair, and safe is paramount. Human oversight cannot be overlooked. Doing so would mean letting AI systems run our lives with neither them nor us knowing what’s going on—inside them or outside.
Therefore, rather than full automation, responsible adoption of generative AI entails keeping humans firmly in the loop. By coupling human creativity and ethics with AI’s scalable capacity, generative systems can be safely harnessed to amplify human capabilities for years to come.