Google

Google, one of the world’s largest technology companies and a wholly-owned subsidiary of Alphabet, took an important step in combating misinformation among its products by investing $6.5 million in the fight against misinformation.¹ This funding will go toward fact-checkers, news organizations, and nonprofits around the world.² The money will also help certain news outlets expose and track coronavirus misinformation.³ Sundar Pichai, CEO of Alphabet and Google, also announced that Google’s Trust and Safety team was working across the globe to safeguard users from phishing, conspiracy theories, malware, and misinformation, and is regularly on the lookout for new threats.⁴

Google Search, the platform’s search engine, is the most popular search engine worldwide with a recent statistic illustrating an 87.35 percent market share.⁵ As the year has progressed, COVID-19 has become the most searched topic, surpassing even some of the most common and consistent queries found in Search with questions such as “Is sneezing a sign of coronavirus?” and “What is PPE for coronavirus” dominating the search engine.⁶ Given the size and reach of this search engine, it is critical for the platform to combat misleading information.

Prior to the spread of the coronavirus, Google published a white paper in February 2019 describing its three strategies for dealing with misinformation across all of its products, including Search, by stating that they “make quality count in our ranking systems, counteract malicious actors, and give users more context.”⁷ Specifically, Google stated that it uses ranking algorithms to organize search results. These ranking algorithms are designed to surface content that the platform determines is high-quality and relevant to a user’s query. Although there has been some contention that the company uses its ranking algorithms to provide preference to content that aligns with certain ideological viewpoints,⁸ the company has stated that Search is designed to make determinations about the usefulness and relevance of a webpage based on a range of signals, and not to promote the political and ideological perspectives of the individuals who built or audited the system.⁹ When Search’s ranking algorithms identify content as misinformation, they will downrank that content so that it appears lower down in search results.¹⁰

Google employs other tools to combat misinformation including human review systems and systems that can reduce spam activity at scale.¹¹ Google says its algorithms can detect the majority of spam and automatically prevent the ranking system from promoting such content by demoting or removing these webpages.¹² The remainder of spam results are typically manually addressed by a spam removal team.¹³ They review the pages in question, typically based on user feedback, and flag them for penalty if they have been found to violate the webmaster guidelines.¹⁴ Manual actions can be used to penalize an entire website, subdomain, sections of a website, or specific pages.¹⁵ Manual action can also demote websites in search rankings and delist them.¹⁶ However, it is unclear how much manual action is taken in response to COVID-19 misinformation.

Finally, for some content, Google provides more context to users through mechanisms such as knowledge panels that connect a variety of sources on a topic, fact-check labels that illustrate verified information, and feedback buttons that directly send information to Google.¹⁷ These mechanisms play an important role in the search engine to help inform the user. Knowledge panels are information boxes that appear on Google when a user searches for entities (people, places, organizations, things) that are in the Knowledge Graph.¹⁸ The Knowledge Graph is the engine that powers the panel.¹⁹ Specifically, the Knowledge Graph is Google’s systematic way of putting facts, people, and places together to create interconnected search results that they determine are more accurate and relevant.²⁰ Knowledge panels are automatically generated, and information that appears in a knowledge panel comes from various sources across the web.²¹

Fact-check labels are appended to articles that include information fact-checked by news publishers and fact-checking organizations.²² These labels appear when a user conducts a search on Google that returns an authoritative result containing fact-checks for one or more public claims.²³ When this occurs, the user will see that information clearly on the search results page.²⁴ The snippet will display information on the claim, who made the claim, and the fact-check of that particular claim.²⁵ However, this information isn’t available for every search result, and there may be search result pages where different publishers checked the same claim and reached different conclusions.²⁶

After COVID-19 was declared a public health emergency by the WHO in late January of 2020, Google launched an SOS Alert with resources and safety information from the WHO.²⁷ Additionally, Google worked with relevant agencies and authorities in the United States to roll out a website focused on education, prevention, and local resources.²⁸

Google has also taken major steps to prevent the spread of misinformation across its other products. For example, the platform announced it was blocking all ads in its Google Ads service that capitalize on the coronavirus.²⁹ Google Ads is the primary mechanism through which businesses can deliver and place ads on Google.

In addition, Google updated their Inappropriate Content policy so that it amended their “Sensitive Events” category. The policy now prohibits acts:³⁰

“Appearing to profit from a tragic event with no discernible benefit to users; price gouging or artificially inflating prices that prohibits/limits access to vital supplies; sale of products or services (such as personal protective equipment) which may be insufficient for the demand during a sensitive event; using keywords related to a sensitive event to attempt to gain additional traffic.”

This policy also includes a ban on the sale of face masks.³¹ Further, Google Play has begun prohibiting developers from capitalizing on sensitive events, in addition to enforcing their long-standing content policies that strictly prohibit apps featuring medical or health-related content that is misleading or potentially harmful. Additionally, Neil Kumaran, product manager for Gmail Security, and Sam Lugani, lead security of product marketing management for G Suite and Google Cloud, shared some steps for administrators to effectively deal with the rising tide of spam emails, and detailing best practices for users to avoid threats.³² According to them, Gmail blocked more than 240 million coronavirus-related spam messages.³³

Given the massive changes occurring across Google’s many products, the company must provide greater transparency and accountability around its COVID-19-related efforts. Specifically, the company should provide periodic updates to consumers on a number of data points, including how many ads have been rejected and removed, how many COVID-19 misinformation-related search results have been downranked or removed, how many spam emails related to the pandemic have been blocked, and statistics on the types of searches made over the course of the global crisis (i.e. searches for specific treatments mentioned by policymakers, searches about the origin of the virus, and searches for critical products). The latter will allow researchers and civil society to better understand information-flow stemming from the virus. Following the pandemic, Google should also publish a comprehensive report on these factors. In addition, where appropriate, the FTC should hold businesses and sellers who run online ad campaigns accountable, by enforcing Section (5)(a) of the FTC Act against any who engage in unfair and deceptive practices.