Nine ways to avoid falling victim to AI washing

In the past few months, artificial intelligence has been a buzzword that everyone talks about. Silicon Valley startups and Fortune 500 companies alike are seeing AI revolutionize industries as the pace of AI steadily accelerates. But excitement, progress and red flags like AI washing are booming at the same rate. Despite the fact that some companies have minimal or non-existent AI adoption, they are exaggerating their AI capabilities because they are eager to jump on the money train and cash in on the hype.

This marketing strategy, while problematic, can help them secure larger seed, Series A, and Series B funding rounds than non-AI startups. According to GlobalData, AI startups raised more than $50 billion in venture capital funding last year alone, and that number is expected to grow this year given the frenzy surrounding ChatGPT and other companies.

Given the capital pouring into these startups, the AI ​​washing phenomenon will only become more intense. The US Federal Trade Commission (FTC) is fully aware of this danger and has warned vendors to be transparent and honest when advertising their AI capabilities.

Michael Atleson, an attorney in the FTC's advertising practice, wrote in a blog post: "Some products that claim to have artificial intelligence may not even have the functionality advertised. In some cases, regardless of the product, Whatever other harm may be caused, this kind of exaggerated propaganda may exist. Marketers know that making false claims about product efficacy or making unverifiable claims is the way to survive in this industry, and this is why the FTC enforces the law.”

In this complex situation, it can be very difficult to distinguish legitimate artificial intelligence solutions from marketing gimmicks.

Beena Ammanath, executive director of the Deloitte Global Artificial Intelligence Institute, said: “Companies need to adopt a healthy dose of skepticism when faced with vendor claims about their AI products. As with anything, if this sounds like It sounds too good to be true, so that's probably what it is."

Donald Welch is the CIO of New York University. If CIOs and their companies don't find the right answers, he said, the consequences they could face include failed or delayed projects, financial losses, legal cases, reputational risk, and ultimately being fired. "I've actually seen executives get fired over this, and I can't even say it was a bad decision." Fortunately, there are several strategies they can use to avoid mistakes.

AI-driven businesses need skilled employees Vetting businesses claiming to use AI can be a long and time-consuming process process. However, there are simple ways, such as searching on LinkedIn, to uncover valuable information for evaluating the business.

Ammanath said: “Video the level of AI experience and education the vendor’s employees have. Companies that are developing AI solutions should have talent in this area, which means they must have data scientists and data scientists. Engineers, and these engineers have deep experience in artificial intelligence, machine learning, algorithm development, etc.”

In addition to looking at employees, CIOs can also look for opportunities to work with external artificial intelligence experts and research institutions evidence. This category includes partnerships with universities, participation in industry conferences and events, and contributions to open source AI initiatives. It's also a good sign if the vendor has experience with similar projects or applications, as this suggests it can deliver high-quality results.

"Double-check the supplier's history," says Vira Tkachenko, the Ukrainian-American chief technology and innovation officer at Ukrainian-American startup MacPaw. If If a company is an expert in artificial intelligence, then it is likely to have a history of publishing research papers in this field or other artificial intelligence products.”

Look for a well-designed data strategyCompanies that truly integrate AI into their products also need a well-designed data strategy, because AI algorithms require it. They need to work with high-quality data, and the more generous and relevant the data, the better the results.

Ammanath said: "AI systems are driven by large amounts of data, so these companies should also have a carefully constructed data strategy and be able to explain how much data is collected and from what sources."

Another issue to watch is whether these companies put enough effort into complying with regulatory requirements and maintaining high data privacy and security standards. With the rise of data privacy regulations such as the General Data Protection Regulation (EU GDPR) and the California Consumer Privacy Act (CCPA), organizations must be transparent about their data practices and provide individuals with control over their personal data . If that's not the case, that's a red flag.

Ask for evidence to support your claims

While the slogan may be tempting, asking for evidence calmly can be helpful. “Asking the right questions and asking for evidence of a product’s functionality is critical to peeling back the marketing and sales rhetoric to determine whether a product is truly powered by AI,” Ammanath said.

Need to evaluate what appears to be a product When considering a specific product or service powered by AI, CIOs can ask how the model was trained, what algorithms were used, and how the AI ​​system will adapt to new data.

Tkachenko said: "You should ask the vendor what libraries or AI models they use. They may simply call the OpenAI API to implement their own functions."

Management and technology consulting firm BearingPoint Partner and global technology leader Matthias Roeser agrees. He added that the components and framework should be thoroughly understood, and the assessment should include "ethics, bias, feasibility, intellectual property and sustainability."

This survey can help CIOs more fully understand the product's Real features and limitations to help them decide whether to buy the product.

Focus on Startups

Startups position themselves at the forefront of innovation. However, while many of them are pushing the limits of the field of artificial intelligence, some may simply be exaggerating their abilities to gain attention and money.

Vlad Pranskevičius, co-founder and chief technology officer of Ukrainian-American startup Claid.ai, said by Let's Enhance: "As the CTO of a machine learning company, I often encounter cases of AI washing, especially in The startup community.” And, he notes, it’s gotten more serious lately. This phenomenon is especially dangerous during a hype cycle like the current one, where artificial intelligence is viewed as a new gold rush, he added.

However, Pranskevičius believes that in the near future, as regulations surrounding artificial intelligence become more stringent, AI washing will be controlled.

Build a reputation for technical expertise

It’s not uncommon for a company to acquire dubious AI solutions, and in this case, the CIO doesn’t There must be something wrong. This may be "the result of poor corporate leadership," Welch said: "The company fell for the marketing hype and overrode the IT team, who was left to pick up the pieces."

To prevent things like this At this time, companies need to cultivate a collaborative culture in which the opinions of technical experts are valued and their perspectives are fully exposed.

At the same time, CIOs and technology teams should build their reputations within the company so that their input is more easily incorporated into the decision-making process. To achieve this, they should demonstrate expertise, professionalism and soft skills.

Max Kovtun, chief innovation officer at Sigma Software Group, said: "I don't think it will be difficult for the CIO to review AI washing. The bigger issue may be business stakeholders or entrepreneurs pushing the use of artificial intelligence in any form, Because they want to appear innovative and cutting-edge. So the right question is how to avoid throwing yourself into AI washing under the pressure of starting a business."

Beyond Buzzwords

When comparing products and services, you must evaluate them with an open mind and see their attributes thoroughly.

Tkachenko said: "If the only advantage of a product or service for you is artificial intelligence, you should think twice before purchasing. It is best to research its value proposition and features, and only if you understand Welch agreed and asked, "Would I buy a system because they wrote it in C, C, or Java? As me As part of the due diligence, I might want to find out if they can maintain the code, company viability, etc."

Doing a thorough assessment may help businesses determine if the product or service they plan to purchase is consistent with their goals Consistently, is it possible to provide the expected results.

Kovtun said: "The more complex the technology, the harder it is for non-experts to understand it and the harder it is to verify that the application of the technology is correct and meaningful. If you have decided to leverage artificial intelligence technology for your company, you'd better Hire knowledgeable experts who have experience in the field of artificial intelligence. Otherwise, your efforts may not bring the benefits you expect."

Follow artificial intelligence related news

Keeping informed about AI-related products and issues surrounding them can also help CIOs make informed decisions. This way, they can identify mistakes they may have made while also taking advantage of new ideas and techniques.

Art Thompson is CIO for the City of Detroit. "I don't think there's enough education right now," he said, advising CIOs to do enough research to avoid falling into the trap of a new or experimental technology that promises more than it can deliver. If that happens, "the time it takes to reprice and replace products can really hurt employees and make them less able to support any changes," he said. “Not to mention the difficulty of people investing time to learn new technologies.”

Additionally, staying up to date on the latest AI-related matters can help CIOs anticipate regulatory changes and emerging industry standards, which can help them comply and stay Competitive Advantage.

And it’s not just the CIO who needs to stay current. “Educate your team or hire experts to add relevant capabilities to your portfolio,” said BearingPoint’s Roeser.

Additional Regulatory Action Around Artificial Intelligence

Upcoming new regulations could simplify the CIO’s task of determining whether a product or service incorporates true artificial intelligence technology. The White House recently released an Artificial Intelligence Bill of Rights, which includes guidelines for responsible design of artificial intelligence systems. In the coming years, more regulations may be introduced.

Ammanath said: “The premise behind these actions is to protect consumer rights and humanity from the potential harm of technology. We need to anticipate the potential negative impacts of technology to mitigate risks.”

Ethics should not be just an afterthought

Businesses tend to influence discussions of new technologies, emphasizing potential benefits while often downplaying potential negative impacts.

Philip Di Salvo, a postdoctoral researcher at the University of St. Gallen in Switzerland, said: "When a technology becomes a buzzword, we often lose focus on the potentially harmful effects it can have in society. Research shows that companies is driving the discourse around artificial intelligence, and technological determinism remains dominant."

This belief that technology is the primary driver behind social and cultural change can overshadow discussions around ethical and political implications , leaning towards a more marketing-oriented argument. As Di Salvo puts it, this creates “a fog of argument that makes these technologies and their producers even more obscure and irresponsible.”

To address this issue, he points to a key challenge in communicating to the public what AI actually is not and what it cannot do.

Di Salvo said: "Most of the artificial intelligence applications we see today, including ChatGPT, are basically built around large-scale statistics and data analysis applications. This sounds possible is a boring definition, but it helps avoid any misunderstandings about what 'intelligence' in the definition of 'artificial intelligence' refers to. We need to focus on real issues like bias, social sorting and other issues rather than hypothetical, speculative long-term assumptions about sex.”

Source: www.cio.com

The above is the detailed content of Nine ways to avoid falling victim to AI washing. For more information, please follow other related articles on the PHP Chinese website!