The Barriers and Opportunities for Generative AI in Corporate WorkplacesCorporations are understandably wary of AI tools like ChatGPT, but it's possible to implement them successfully and safely.

Companies of all sizes have become accustomed to using predictive AI to achieve a range of outcomes, such as anticipating risk, developing new products and forecasting buying behaviors. However, many enterprises are struggling to figure out how to realistically incorporategenerative AIinto their operations. It poses many advantages, of course, but it's also fraught with fear and uncertainty.

Perhaps because of that, only12% of IT decision-makersrecently surveyed by Enterprise Technology Research, as reported by the Wall Street Journal, said they plan to use OpenAI technology — creator of the most popular generative AI tool, ChatGPT. Yet, the global generative AI market is expected to reach$111 billion by 2030, per Acumen Research and Consulting.

With all the buzz around it and advancements in the technology, there's little doubt that generative AI is going to be an asset across industries as widespread as healthcare, insurance and logistics. However, it's a newer solution. As such, businesses and their leadership teams are only starting to determine how best to leverage it to its fullest — and safest — degree.

This leaves corporate leaders at a crossroads. Many want to bring generative AI solutions in-house. Some — particularly those at enterprise-level corporations — have even put a budget behind this desire. They want to access this emerging technology in the most efficient ways possible. I believe the easiest way to make that happen is for businesses to join forces with AI-based startups.

Related:The Secret to How Businesses Can Fully Harness the Power of AI

Attributes, advantages and areas of concern around generative AI

Because of its continual learning capacity, generative AI might well be described as creative AI. That is, it cancreate content以前不存在的。虽然这是令人兴奋的,我t's brought about much discussion on how to handle its downsides, such as inaccuracies. Generative AI isn't able to identify or self-correct when it gets things wrong or even pushes out content that's inappropriate or biased.

Another overarching problem with generative AI concerns data. Because it's trained on vast amounts of data, it may produce content that violates intellectual property rights. What is the law around generative AI content that leans heavily on existing content? It's a fine line between unique expression and plagiarism, and the laws haven't quite caught up to where that line lies.

In addition, vertical, industry-specific solutions with unique data libraries, rather than general generative AI models, provide the most applicable answers but can be costly. Accessing the vast amounts of data needed to produce accurate insights is expensive, and the computing power required to do so is highly demanding and unsustainable in terms of expense. However,Microsoft seems to be exploring collaborations with AMDto lower computing costs, and potential software technologies could reduce computing consumption.

Of course, generative AI is far from being all negatives and no positives. Due to its transformative nature as a technology, it could become a tool for sector disruption, helping companies save time and resources and improve their decision-making.

In my view, I see generative AI as a value-added tool that's only going to become more capable and intelligent. New models are emerging that could address the issues of cost by using smaller data sets, but it will take a few years for new models to evolve to a stage where they are affordable and user-friendly enough for practical applications. At present, generative AI is most effective when used inconjunction with human input. Human intervention fosters consideration of different perspectives and minimizes ethical and flawed data risks.

TakeChatGPT, for example. The quality of its output and answers depends on the quality of the input and human intelligence involved. To get high-quality answers, content and results from ChatGPT, human users must take active roles in the process to create feedback loops. Otherwise, ChatGPT (and similar generative AI solutions) is interesting but not reliable or holistically useful.

Related:The Top Fears and Dangers of Generative AI — and What to Do About Them

Collaboration: Key to bringing generative AI solutions into corporate settings

Collaboration between startups and corporate enterprises can be the game-changing factor across the entire generative AI landscape. Not only do partnerships allow founders to explore various options and even work with different model providers, but they also lower the barriers for companies to access generative AI. It also produces more interest in open-source model ecosystems. With open-source contributions, there can be a collective and effective effort to push generative AI's boundaries, challenge dominant AI players and drive down costs. Ultimately, it fuels a positiveinnovation environmentfor both the startup and the collaborating corporation.

Collaboration offers another opportunity: Businesses and generative AI solutions startups can focus on implementation and adoption rather than investing in more fundamental systems. Such a partnership will entice large companies to integrate generative AI into their workflows, making it less complicated for the startup to explore faster and potentially attract more investors for future developments.

With that being said, enterprises won't just center into a partnership with a generative AI startup without consideration. Keep these things in mind to streamline and inform your decision-making when partnering:

1.首席信息官和首席技术官必须适应the solution

Right now, CIOs and CTOs are in a state of panic. Why? They're being pressured by their boards to understand the implications of generative AI because it accesses sensitive data. Consequently, although partnering with a startup is a perfect way to train and retrain a generative AI model with industry-specific input to ensure accuracy and consistency, it may feel like a liability risk.

To help the CIO and CTO get comfortable, talk about whatdata securitymeasures are or could be put into place. This could include data encryption solutions and secure learning techniques. Once these measures are established, the major players in your business are likely to be more confident about implementing generative AI internally. Remember: Most CIO and CTO executives understand that generative AI will need domain knowledge and access to unique industry data libraries. They simply want to avoid a breach that could put your brand in an unwanted spotlight.

Related:Generative AI: the Rising Kid on the Start-up Block

2. The employees will have to learn how to effectively use generative AI

If you want employees to jump in and implement generative AI for thatcompetitive advantage提升,你必须让它发生。这意味着更多的than just implementing generative AI applications. It means explaining the best practices regarding the technology's use and data regulations. Presently, there are extensive discussions swirling around data regulation, so your team will need to stay up to date.

Providing the most current information on the regulation of the usage and processing of data — not to mention data ownership concerns — to employees is critical. The more they know, the more they can control their generative AI usage andmitigate problems.

Generative AI is making a huge splash across the world right now, especially with last year's release of ChatGPT. While it's still in its infancy, corporations such as yours can get ahead of the pack by working with startups developing generative AI models and applications. You just need to conduct some due diligence to ensure you get all the advantages of generative AI and avoid preventable snags.