Microsoft is building out A.I., but here's what it thinks could go wrong

  • Microsoft has been focusing more on generating revenue from the artificial intelligence and the internet of things categories.
  • The company said social media could amplify issues that arise.
Microsoft CEO Satya Nadella at a company event in New York in May 2017.
Source: Jason DeCrow, AP Images | Microsoft
Microsoft CEO Satya Nadella at a company event in New York in May 2017.

Microsoft pointed to new risk factors connected to markets it's going after in its latest annual report, which came out on Friday.

The company has warned investors before about the risks involved in cloud services, where it has become a bigger force. Now, as Microsoft seeks to execute on growth opportunities associated with artificial intelligence and devices that connect to the internet, it's flagging potential issues that could arise in those areas.

While Microsoft doesn't acknowledge the dustup over its work with U.S. Immigration and Customs Enforcement in its report for the year that ended on June 30, it does suggest there is the potential for "brand or reputational harm" that could arise from the use of certain AI technologies:

Issues in the use of artificial intelligence in our offerings may result in reputational harm or liability. We are building AI into many of our offerings and we expect this element of our business to grow. We envision a future in which AI operating in our devices, applications, and the cloud helps our customers be more productive in their work and personal lives. As with many disruptive innovations, AI presents risks and challenges that could affect its adoption, and therefore our business. AI algorithms may be flawed. Datasets may be insufficient or contain biased information. Inappropriate or controversial data practices by Microsoft or others could impair the acceptance of AI solutions. These deficiencies could undermine the decisions, predictions, or analysis AI applications produce, subjecting us to competitive harm, legal liability, and brand or reputational harm. Some AI scenarios present ethical issues. If we enable or offer AI solutions that are controversial because of their impact on human rights, privacy, employment, or other social issues, we may experience brand or reputational harm.

This isn't the very first time Microsoft is acknowledging that it could run into some problems in connection with AI. In its most recent quarterly earnings report, the company said its actual results could differ from guidance because of "issues about the use of artificial intelligence in our offerings that may result in competitive harm, legal liability, or reputational harm," as VentureBeat reported.

Even if Microsoft said the ICE work didn't involve AI, the controversy over that business relationship became inflamed as people discussed it on Twitter and other social channels, and indeed, in this year's annual report Microsoft says that social media could worsen the impact of issues:

The proliferation of social media may increase the likelihood, speed, and magnitude of negative brand events. If our brands or reputation are damaged, it could negatively impact our revenues or margins, or ability to attract the most highly qualified employees.

Microsoft also acknowledged the risks that could develop as it does more with the so-called internet of things, including in the realm of personal security:

The development of the internet of things presents security, privacy, and execution risks. To support the growth of the intelligent cloud and the intelligent edge, we are developing products, services, and technologies to power the IoT, a network of distributed and interconnected devices employing sensors, data, and computing capabilities including AI. The IoT's great potential also carries substantial risks. IoT products and services may contain defects in design, manufacture, or operation, that make them insecure or ineffective for their intended purposes. An IoT solution has multiple layers of hardware, sensors, processors, software, and firmware, several of which we may not develop or control. Each layer, including the weakest layer, can impact the security of the whole system. Many IoT devices have limited interfaces and ability to be updated or patched. IoT solutions may collect large amounts of data, and our handling of IoT data may not satisfy customers or regulatory requirements. IoT scenarios may increasingly affect personal health and safety. If IoT solutions that include our technologies do not work as intended, violate the law, or harm individuals or businesses, we may be subject to legal claims or enforcement actions. These risks, if realized, may increase our costs, damage our reputation or brands, or negatively impact our revenues or margins.

Microsoft also suggested that it could face challenges because of changes in trade policies, like tariffs, which U.S. President Donald Trump has sought to impose. The company has previously talked about how tariffs can affect it, but this year it goes into more detail:

Changes to trade policy or agreements as a result of populism, protectionism, or economic nationalism may result in higher tariffs, local sourcing initiatives, or other developments that make it more difficult to sell our products in foreign countries. Disruptions of these kinds in developed or emerging markets could negatively impact demand for our products and services or increase operating costs.

Another interesting change: San Francisco start-up Slack, whose app lets team members set up chat rooms, is mentioned as an Office competitor for the first time in Friday's filing. Microsoft recently introduced a free tier of its Slack competitor, Teams.