Artificial intelligence is already changing the way the world does business in a number of fundamental ways, but the extent to which businesses embrace AI will depend on the sector in which they are operating. For some industries, AI is becoming increasingly critical. For example, in the investment management and insurance sectors, companies are already using AI to monitor financial markets and external events in an effort to get a better grip on quantifiable risk.
A great deal is being written about the modern approach to the many applications of artificial intelligence, but it is important to understand its limitations as well. Ultimately, AI and machine learning depend on good data, in large quantities. AI is really only as good as the data it analyzes and consumes. If a company has access to large quantities of data, there is scope for looking at how AI can make business intelligence more efficient and more competitive – but if the data isn’t sufficient, accurate and relevant, the results may be less than impressive.
This is why some groups are exploring ways of making third-party data available to businesses for use within their own systems, from CRM to risk management. The client firms may not be able to actually see and process that data themselves, but using the facility of the cloud, on-premises AI functions can also benefit from the wider learning inputs available.
Open-Source Artificial Intelligence
The Linux Foundation, a driving force in the promotion of open-source software and solutions, is creating considerable interest within the international business community because it has espoused an open-source approach to the development of artificial intelligence that may prove to be of great value. As part of its LF Deep Learning project, Linux has launched the Acumos AI Project, a comprehensive platform for AI model discovery, development and learning.
Acumos aims to help businesses by promoting a high level of collaboration between them for the development of AI science. Among the groups already involved in the project are Chinese giants Tencent and Baidu, as well as AT&T and Tech Mahindra.
“AI is already transforming business models and delivering contextual customer engagements, creating new opportunities for service providers, content developers and other industry players to build the enterprise applications that will run our increasingly digital society,” says Anthony Goonetilleke, group president of software and service provider Amdocs, which is also a contributor to the project.
The open-source movement is important because it enables business to leverage the artificial intelligence findings and data of other businesses. A joint project like Acumos can help participants large and small realize benefits from AI that they could not do if they were pursuing the technology on a standalone basis.
Amdocs says it has been researching AI for years, with a particular focus on the development of personalized digital engagements, but open source collaboration can still confer advantages. Says Goonetilleke: “What makes Acumos so exciting is that industry leaders will bring innovative, data-driven insights and technology from global sources to empower experiences with cutting-edge AI-enabled capabilities.”
Identifying Patterns in Diverse Data
AI can also help companies to make the most of different data sets, identifying and isolating important patterns. For example, if a business is regularly impacted by the weather, adding third-party weather data to its analytics can enable executives to make more informed decisions based on forecasts. At the simplest level, for a provider of ingredients for ice cream manufacturers, weather and sales data combined can make it easier to evaluate likely demand, and respond accordingly.
Where most businesses will see AI at work on a practical level will be within their customer relationship management system. Platforms such as Salesforce’s Einstein and others are already leveraging AI marketing technology to help make sales and marketing teams more efficient. By pooling learning from literally millions of CRM users, they can deliver insights to individual sales teams that would not be achievable using the customer data of a single organization.
AI within the CRM process can help to inform and generate sales activity based on customer data and behavior patterns as well as automating many of the more time-consuming activities surrounding the sales process.
Will Employees be Replaced by Robots?
However, concern persists that artificial intelligence in business will eventually replace human beings, putting thousands or millions of people out of work, substituting them with AI robots. There is certainly a risk to specific job functions, but overall the threat is probably somewhat over-stated. In all likelihood, artificial intelligence will lead to changes in working habits and patterns, just as the original industrial revolution precipitated a shift in labor from farms to factories.
The purpose of AI in most workplaces is ultimately to make employees more efficient, not to replace them altogether. In some companies employees find themselves spending two hours or more of their working week putting together reports. Within the social work sector, this proportion can be much higher due to the increased requirements placed upon employees by government bureaucracy and the demands of increased public accountability.
In the customer relationship context, many sales people spend a large part of their day – Salesforce estimates as much as 20% – entering or editing data in CRM systems. Employers would prefer that this time is spent actually selling and generating revenue rather than maintaining records or timesheets. Part of the reasoning that underpins the creation of Salesforce’s Einstein was to make salespeople more efficient, freeing them to spend more of their working time on core tasks.
Execution: Clarity of Vision
This kind of efficiency gain is set to be repeated in many other areas of business. But what are the best ways to implement AI within a company?
First business leaders need to establish clear business goals – what do they want their business to achieve? What tasks do they need to see addressed? Answering these questions can enable a pilot project to be established to examine what systems and technology can be brought to bear on the issues raised. Ideally this should be overseen by a senior executive within the organization. There may be a temptation to hand this over to the head of IT, but if practical results are desired, a more senior manager should be involved.
Next comes identifying the data sets and resources that any AI project should draw on, arguably the most crucial part of the project. This step will create a foundation for any testing and improvements as the initiative progresses. Executives should expect to conduct plenty of fine-tuning of algorithms before they achieve final results that will benefit the company.
Artificial intelligence is not going away – most large businesses already have a strategy in place. Even for mid-sized firms, it is likely that one or more of their direct competitors are already looking at the future potential of the countless applications of AI. There’s no time to be lost for businesses that don’t want to be left behind.
link
More Stories
Innovation through collaboration? Take care of your patent rights
Collaborating with EPA through the Federal Technology Transfer Act
Collaboration Patterns between Owners, AECs and Environmental Consulting Firms