AI in business intelligence: Caveat emptor

One of the ways in which organizations use the latest artificial intelligence algorithms to help them grow and prosper is to adopt special artificial intelligence models to harmonize their business strategies.

The distinction between private and public artificial intelligence is important in this context – most institutions are truly cautious against allowing access to sensitive data groups, such as human resource information, financial data and the details of the operating date.

It is logical that if artificial intelligence is granted specific data that can be built its responses, then its result will be more important, and therefore it is more effective in helping decision makers to judge how the strategy. The use of private thinking engines is the logical way that companies can get the best results from artificial intelligence and maintain intellectual property safe.

Data for institutions and the ability to control the local artificial intelligence model give organizations the ability to provide allocated prediction and operational control that is mostly based on the daily reality of the company’s work. The Instrage Institution Paper Paper Paper Paper Paper is launched, and puts the use of internal data as a competitive feature, and the Accene AIS describes as “ready to provide economic rise and change work since agricultural and industrial revolutions.”

However, there is a possibility that it will be like traditional business intelligence, using historical data derived from several years of operations throughout the institution, to consolidate decision -making in patterns of the past. McKinsey says that companies are in danger “the opposite of their institutional past in the algorithm.” Harvard Business Review captures some artistic complexity, saying that the act of allocating a model so that the company’s more related activities are difficult, and therefore, not the task that must be taken by anything but the most dirty at the level of data science and programming.

MIT Sloane achieves a balance between enthusiastic defenders and the conservative voices of special males in the work strategy. Artificial intelligence is recommended to be considered a participant pilot, and urges to continue interrogation and verify artificial intelligence, especially when the risks are high.

It believes in the revolution

However, decision makers who think about following this work (obtaining the wave of artificial intelligence, but doing this in a special and safe way) may want to consider the motives of the sources of advice that strongly calls for enabling artificial intelligence in this way.

Deloitte, for example, adopts and manages AI’s solutions for customers who use customized infrastructure such as factory offers as a service, while Acceneture has a special practices for AI’s strategy for its customers, such as Acceneture Applied Intelligence. It gets used to AWS and Azure, and build artificial intelligence systems for Fortune 500 companies, among other things, Deloitte partners with Oracle and NVIDIA.

With “Skin in the Game”, phrases like “the most important […] The change in work since the agricultural and industrial revolutions “and” detailed compass “is inspiring, but the motives of the sellers may not be completely altruism.

Artificial intelligence defenders generally indicate the ability of models to determine statistical trends and more efficiently human trends. Looking at the data block available for the modern institution, which includes both internal information available abroad, the presence of programs that can analyze data on a large scale is an incredible feature. Instead of creating a manual analysis of the huge data warehouses-which take a long time and reactive of errors-you can see artificial intelligence through peeling and real impregnated visions.

Ask the correct questions

In addition, artificial intelligence models can explain the sofa in the regular language, and to provide predictions based on experimental information, which, in the private AIS context, is very related to the institution. The relatively non -skilled employees can inquire about data without having skills in statistical analysis or information on databases, and obtaining answers that have included several teams and skills groups derived from all over the institution. Saving time alone, which allows institutions to focus on the strategy, rather than forming the necessary data points and manually inquiring the information they were able to collect.

Both McKinsey and Gartner warn, however, of excessive confidence and limitations in data. In the latter, historical data may not be related to the strategy, especially if the records return for several years. It may be better to describe excessive confidence in the context of artificial intelligence as factors that trust in artificial intelligence responses without question, and not independently in the details of the responses, or in some cases, taking into account the responses to the queries that were depicted badly.

For any software algorithm, human phrases such as “the rule of results that have reached our historical data” open to interpretation, unlike, for example, “establishing the results reached on sales data in the past twelve months, ignoring extremist values ​​that differ from the center by more than 30 %, although these are counterparts to take into account.”

Experience program

Organizations may follow Amnesty International Special Solutions along with mature and current business intelligence platforms. SAP business institutions are nearly 30 years old, but a young man compared to the intelligence of SAS, who has been present since the Internet, has become prevalent in the 1990s. Even the new arrivals such as Microsoft Power Bi represent at least a decade of development, repetition, customer comments and the use of the real world in business analysis. Therefore, it seems logical that the spread of special artificial intelligence on work data in addition to the strategic tool group, instead of a silver bullet that replaces “traditional” tools.

For private artificial intelligence users who have the ability to review and adjust their style inputs and internal algorithms, retaining control of humans and supervision is important – just as it is with tools such as Oracle Business Intelligence Suite. There are some scenarios that give smart data processing in actual time (online retail mechanisms, for example) is a competitive analysis on current BI platforms. But Amnesty International has not yet developed into a Swiss Swiss knife for the strategy of work.

Until Amnesty International, which is designed to analyze business data, repetition, stiff battle, and mature like some BI platforms on the market, may reduce the first trap of the enthusiasm of AI and AI sellers with practical experience and critical eye. Artificial intelligence is a new tool, one with a large amount of capabilities. However, it remains the first generation in its current body, public and private.

(Image source: “It relates to the rules and strategy” by PSHTTERBG licensed under CC by 2.0.))