dcsimg
 

8 AI Implementation Risk Factors For Small Business

Implementing artificial intelligence can cause risks for small businesses. Eight business leaders give recommendations for small businesses that are considering AI.

Artificial intelligence, commonly referred to as AI, is a widely known concept that surrounds our day to day lives. Whether it be Google auto-filling your search query or Apple Maps generating your ETA, examples of AI are constantly around us. These services not only improve our day to day lives but also create opportunities for businesses to operate more efficiently.

If you clicked on this article, chances are you probably know that for all the opportunities AI presents, it also introduces risk. To get a better understanding of what these risks are and what they mean for small businesses we interviewed eight business leaders to get their take. Keep reading to learn the lay of the land before implementing any form of artificial intelligence in your organization.

Misleading Input of Information and Data

AI relies on an input of data to produce key insights. An industry like healthcare can use AI to comb patient records to develop healthcare trends, or a tech company may have a chatbot that relies on AI to produce responses. The risk is when the data pulls may be misleading. If there’s errors in the patient medical records, then the AI recommendations made to doctors  and healthcare insurance companies could seriously impact patient lives. Or in the case of Microsoft’s Tay, an influx of tweets from racially charged Twitter users caused Tay to post inflammatory and offensive tweets and be shut down just 16 hours after launch. The input of data has to have integrity for the output of AI to be reliable. 

Brett Farmiloe, Markitors

Implementation Costs

Any successful AI implementation starts by determining very precisely the business problem that needs to be solved. Owners should have a clear idea about where the problem lies in their business workflow, what KPIs reflect it, and what ROI they expect by using an AI solution to fix it. Framing your approach in terms of performance, current and expected, will help you assess if, and when, investing in AI is right for your business. In addition to the implementation costs, ensuring that your data is available and in the proper format before embarking on this journey will save you time and money throughout the process.

Roger Chahine, Day223 Ventures

Expecting Too Much

Expecting too much from AI could be a potential risk. Artificial intelligence has limitations. Oftentimes small business owners will implement an AI-based software solution with high hopes. Because of the hype surrounding AI, the features within that solution fail to live up to the expectations of a small business owner. To mitigate the risk, lower the expectations. That way AI can comfortably live within limitations and still drive value for a business. 

Megan Chiamos, 365 Cannabis

Absence of Return on Investment

Companies need to invest both time and money to clearly understand the benefits of an AI application. In fact, small businesses should learn from the mistakes of big enterprises to ensure they don't fall in the same trap. Oftentimes large companies make huge investments in AI because they can afford to take that risk. Many companies have a part of their overall budget dedicated to an innovation fund that allows them to use it for such experiments. However, SMEs do not have that luxury so they should be very cautious and should model the best and worst case scenarios for expected returns before making any AI purchase.

Akanksha Shrivastava, Deep North

False Barrier to Entry

AI is still largely misunderstood by some, which puts a false barrier to entry. The automation systems powered by AI still need to be overlooked, maintained, and improved for the best results. Half-hearted attempts at AI implementation end up costly. Small business owners must cooperate with AI experts to ensure seamless integration. We took it step-by-step to make sure every piece is in place before finalizing the AI transformation. Once you get it right, you reap the benefits for years to come. Don't be afraid to invest in AI systems, but make sure you work with experts if you do so.

Zohar Gilad, InstantSearch+

Lack of Data Strategy

The number one barrier is the lack of data strategy and a lack of understanding of how data science can be used. Small businesses usually lack both, and when they try to hire a data scientist to help them out, they might find out that they can't afford it. This is why it is important to get the proper education in AI first, and have a plan in place, before they start thinking about using data science. This is why the Tesseract Academy has been at the forefront of AI education all these years.

Dr. Stylianos Kampakis, Tesseract Academy

False AI Implementation

The biggest risk small businesses have when implementing AI is that they'll implement fake AI. This is software that is sold as AI but is really just a nested stack of conditional statements that are sophisticated enough to look like AI but lack the benefits that SMBs are looking for. Unskilled engineers will sell this software at the AI rate, meaning SMBs will invest a great deal of money on fake AI and not see the ROI that they expect.

The best way to prevent this is to bring in someone skilled in AI to sit in on your meetings with contractors. They will help you ensure that the software you're looking at implementing is real AI. Any consulting fee for this will be well worth making sure you get what you're paying for when it comes to the actual software.

Mark Varnas, Red9

Algorithm Bias

Biases in the algorithm is crucial hence a black box approach to AI is not recommended. Transparency is key in any AI so that customers and internal stakeholder (and if applicable government) understand what is being measured and how. Understanding governance, risk management and ethics is crucial in its deployment.

Bianka Castillo, Recruiting Maven

Terkel creates community-driven content featuring expert insights. Sign up at terkel.io to answer questions and get published. 



This article was originally published on 11-05-2020
eWeek eWeek

Have the latest technology news and resources emailed to you everyday.

Click for a full list of Newsletterssubmit