The rise of new technologies driven by artificial intelligence (AI) has helped level the playing field for new businesses, contributing to the surge in entrepreneurship since 2020. That’s one of the numerous notable findings from Gusto’s annual New Business Formation Report. The company surveyed more than 1,300 business owners who started their companies in 2023.
Gusto found that 22 percent of respondents are using generative AI in their operations. These tools, such as ChatGPT, can generate texts, images, or other content in response to prompts. And 13 percent of survey respondents are either developing new AI technologies themselves or integrating AI directly into one or more features of their products.
Read on to learn more about how the power of AI is affecting entrepreneurship.
AI use cases in business today
From automating routine tasks and deploying customer service chatbots to wielding ChatGPT to generate content, AI is streamlining operations and freeing up new business owners to focus on high-value work, core competencies, and strategic initiatives. Properly implemented, it can cut costs and boost profits across a variety of critical functions.
As Ethan Mollick, a professor at the Wharton School of the University of Pennsylvania who studies entrepreneurship and AI, has put it, “The multiplier on human effort is unprecedented, and potentially disruptive.”
According to Gusto’s report, entrepreneurs are already using AI systems for:
- Legal operations (legal research, contract review): 17%
- Human resources (such as creating job postings): 12%
- Accounting and finance (invoicing, bookkeeping): 11%
- Security and fraud detection (identifying bots, incident response): 5%
By far, though, marketing is the most popular way entrepreneurs are currently taking advantage of AI. Seventy-six percent of the new businesses surveyed that use generative AI are applying those tools to marketing tasks like content creation and market research. A smaller but still significant percentage (41 percent) of new firms are using generative AI for sales (e.g., crafting communications with leads) or customer service work (26 percent).
Among other things, entrepreneurs can employ AI to personalize the customer experience. AI tools can, for example, tailor communications—including emails, product recommendations, and customer service interactions—building and reinforcing customer loyalty along the way.
Mollick ran an experiment recently that demonstrates the far-reaching potential of AI when it comes to marketing. He gave an AI tool 30 minutes to accomplish as much as it could to market the launch of a new educational game. AI did all of the work, with Mollick offering only directions.
He described the results as “superhuman.” In that short span of time, and with fewer than 20 plain English inputs, AI:
- Performed market research,
- Created a positioning document,
- Wrote an email campaign,
- Designed a website, logo, and “hero shot” graphic,
- Made a social media campaign for multiple platforms, and
- Scripted and created an animated video.
The associated savings in time and resources underscores why AI is such a game-changer for entrepreneurs.
Human resources is another area ripe for AI assistance. Startup companies often devote a surprising amount of time to tasks such as drafting employment policies and training materials. Generative AI could assume much of this work, enabling HR staff to direct their energies to more employee- and applicant-facing matters where the human touch is integral. AI can also help write job descriptions and postings, identify passive candidates, devise interview questions, and formulate recruitment and retention strategies.
Potential risks as AI becomes more widespread
For all of its promise, experts caution us to remember Amara’s Law, which states that ”we tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.” Several risks could limit the potential of AI in the future, at least for some entrepreneurs.
Hallucinations
AI can spin inaccurate answers that seem authoritative but are, at best, misleading and, at worst, completely fabricated. In June 2023, for example, two New York attorneys were fined $5,000 for submitting legal briefs written by ChatGPT that included non-existent judicial opinions with fake quotes and citations. The judge in the case found that one of the fake rulings generated by the chatbot had “some traits that are superficially consistent with actual judicial decisions,” but that its legal analysis was “gibberish” and another portion bordered on “nonsensical.”
Transcripts of the attorneys’ interaction with ChatGPT to produce the brief showed the chatbot using words like “sure” and “certainly!” Mollick has described AI as “an intern who wants to make you happy and therefore lies a lot and is kind of naive [and] never admits that they made a mistake.”
Loss of creativity
AI’s eagerness to please can also stifle creativity because the technology may generate “ideas” that are literal or obvious. David Schonthal, a professor of strategy, innovation, and entrepreneurship at Northwestern University’s Kellogg School of Management, told The Wall Street Journal that AI systems will “sense what people say they want, which is often different from what they truly desire.”
For example, an AI system could discover that sales of cheese are rising and, in response, propose creating a new cheese-distribution system instead of “a truly innovative new business.” Schonthal says, “The abstract piece of the process is still a uniquely human capability.”
Moreover, AI models are, by nature, backward-looking. Successful entrepreneurs, on the other hand, need to be forward-thinking.
Legal risks
AI presents a variety of legal risks. For example, generative AI could trigger intellectual property infringement issues. Creators of all kinds have already turned to the courts, alleging unauthorized use of their works by AI systems.
Liability for incorrect information is another risk. In February 2024, for example, an airline in Canada was found liable after its chatbot promised a customer a discount that wasn’t actually available.
AI has also proven vulnerable to bias and discrimination. Amazon had to ditch an AI recruiting engine because it didn’t rate candidates for software developer jobs and other technical posts in a gender-neutral way. It apparently learned to prefer male candidates because the majority of the resumes on which it trained were from men.
Privacy and data ownership
Privacy and data ownership concerns are particularly relevant for increasingly popular large language models (LLM) like ChatGPT, which are trained on vast amounts of data. Users can expose sensitive company information when using such models, providing access to technology companies whose privacy and other standards don’t match up with their own.
Samsung Electronics is one company that fell prey to this problem. Its employees entered confidential source code into ChatGPT to debug the code. They also entered transcripts of internal meetings so the tool could summarize them. By doing so, they assumed the risk that this sensitive data could be used to further train ChatGPT and, at some point, pop up in response to the prompts of other users, potentially including competitors.
One thing is clear for the “age of AI”: Comprehensive cybersecurity now must contemplate these types of risks.
Uncertainty about future regulation
While European governments have already started efforts to regulate AI, the United States lags behind. Some regulation seems likely, though.
AI is a rare new business innovation where business leaders have actively sought additional regulations and global collaboration. In a Deloitte survey of nearly 3,000 business and technology leaders involved in piloting or implementing generative AI in their organizations, 78 percent said that more governmental regulation of AI is needed. Seventy-two percent said there is currently not enough global collaboration to ensure the responsible development of AI-powered systems.
And in May 2023, Sam Altman, the CEO of the company that created ChatGPT, explicitly asked Congress to regulate AI. Appearing before the Senate Judiciary Committee, he supported the creation of a federal oversight agency with licensing authority, as well as the implementation of safety standards and independent audits to test for compliance.
Potential future opportunities
The Deloitte survey also found that most generative AI efforts currently revolve around improving efficiency, increasing productivity, and reducing costs. This tracks with previous adoptions of groundbreaking technologies. Once a business has accomplished these operational goals, though, it’s time to turn to more innovative, strategic, and transformative use of AI.
For instance, businesses can apply AI tools such as big data-driven data analytics to improve their decision-making. Predictive analytics can crunch current and past data to provide actionable insights for everything from financial forecasting to inventory management.
AI systems can dramatically streamline product development, too. In the earliest stages, it can assist business leaders in devising ideas, including assessing the competitive landscape and market conditions by collecting and processing information tucked away in sources as diverse as financial filings, media reports, court documents, governmental agency findings, and social media posts. Once an idea is identified and a prototype developed, AI tools can easily and inexpensively test it.
Lingering reluctance
AI has proved beneficial for many entrepreneurs but hasn’t been universally embraced. Polling shows that numerous hurdles remain for many small businesses.
One apparent reason is the high cost. Small businesses may understandably be wary of the quickly mounting expenses for hardware, software, and infrastructure. Businesses with small staffs and tight budgets might also worry about finding the time and money for training on how best to use AI tools.
Some businesses may just want to avoid all of the risks outlined above. Plus, sticking with the actual employees can improve customer satisfaction. You don’t have to look far to find complaints online from customers frustrated by chatbots.
These businesses might be only delaying the inevitable, though. In the not-too-distant future—if not already—startups that don’t have a plan for adopting generative AI will probably be seen as out of step with today’s business ecosystem. This perception could threaten funding opportunities, as evidenced by the Gusto survey.
It found that, in 2023, 19 percent of new businesses developing AI technologies or using AI in their operations received a private capital investment, compared to just 4 percent of firms that weren’t making or using AI. AI-focused companies also started with considerably more funding: Nearly half of new businesses focused on AI began with more than $10,000 in startup funds, compared to 29 percent of other firms.
Getting started
Dipping your toes into AI waters can seem daunting. Here’s how to reduce some of your risks and fears:
1. Take small bites at first
Tackle simple projects like robotic automation before scaling up organization-wide. Look for repetitive, time-consuming tasks that are vulnerable to human error. To determine which tasks to begin with, consider factors like budget and ease of implementation.
This approach will help you reap some measurable results immediately. You’ll also be able to identify hurdles and gaps that should be addressed before going big with AI technology.
Keep your expectations in line with your initial expenditures, though. Research suggests that entrepreneurs should be prepared to make a significant investment in AI before they see any real gains, as limited AI adoption doesn’t directly contribute to revenue growth. It was only when businesses were using at least a quarter of the AI tools currently available to them that growth rates picked up, and their investments began to pay off.
2. Don’t go it alone
Technical expertise is essential. It seems like new AI tools and systems make a splash on the market every day, making research on pricing, features, customer support, the capacity to integrate with existing systems, and other points vital to choosing the right ones for you.
Of course, many startups don’t have the requisite expertise on staff early on. When costs are a concern, you can start by hiring an independent contractor with the necessary skills. Eventually, you might bring on full-time staff.
3. Remember the human toll
Sixty-five percent of employees report feeling anxious about AI making their jobs obsolete. When you start to adopt AI tools in the workplace, you should anticipate that some of your employees in affected areas will be worried, nervous, and, at the very least, apprehensive about learning how to use the new technologies. You can mitigate these concerns with training and transparency.
The upfront cost of adopting AI tools may seem high, but the investment is likely to be worth it in the long run. New businesses need to keep up with the evolving ecosystem and put themselves in a position to promptly seize new opportunities.
