Facebook chief operating officer Sheryl Sandberg, left, and Twitter CEO Jack Dorsey arrive to a Senate Intelligence Committee hearing on Capitol Hill, Wednesday, Sept. 5, 2018, in Washington. (Jose Luis Magana/AP)

Donald Trump’s allies on social media paved his way to the White House, and today the commander in chief’s Twitter tirades rally his base and set the country’s policy agenda and constitutional debates. His fans admire the alleged authenticity of the president’s direct, seemingly unfiltered communications.

But are the social media masses being duped? Mounting evidence suggests that the president authors only some of his tweets. And those famous campaign phrases and hashtags of “drain the swamp” and “deep state”? According to Chris Wylie, a Cambridge Analytica whistleblower, Stephen K. Bannon directed the testing of these messages in 2014, long before Trump signed up for the presidential race.

Yes, the Kremlin’s manipulation of social media is a threat to U.S. democracy. But some of the most damaging efforts I’ve seen lately are American, not Russian, and they’re far more technically capable than those of the Internet Research Agency that was indicted by special counsel Robert S. Mueller III in February. It’s time we started paying attention to the political campaigns and public-relations firms exploiting social media to drive audiences apart online and constituencies against each other at the ballot box. Western opportunists will adopt the Kremlin’s information warfare art but will apply a more devastating power — that of artificial intelligence — to sway audiences through social media assaults.

Cambridge Analytica’s much-touted use of social media-generated psychographic voter targeting may have been more aspirational than factual, a touch of digital snake oil in the pursuit of clients. But Bannon and Cambridge Analytica were naturally and logically pursuing the next advance in political influence — campaigning that is more science than art, manufactured populism guided by hidden influencers who understand microscopic audience preferences and the psychological vulnerabilities of voters.

Cambridge Analytica’s harvesting of Facebook accounts and pairing with voter profiles merely represents a small first step for social manipulation. Advanced public relations firms, propagandists and campaigns, now and in the future, seek a full digital pattern-of-life on each potential voter. Every like, retweet, share and post on all social media platforms will be merged and matched with purchase histories, credit reports, professional résumés and subscriptions. Fitness tracker data combined with social media activity provides a remarkable window into just when a targeted voter might be most vulnerable to influence, ripe for just the right message or a particular kind of messenger.

Future campaigns will pick not just the issues and slogans a candidate should support, but also the candidate who should champion those issues. Dating apps, the aggregate output of thousands of swipes, provide the perfect physical composite, educational pedigree and professional background for recruiting attractive candidates appealing to specific voting segments across a range of demographics and regions. Even further in the future, temporal trends for different voter blocks might be compared to ancestry, genetic and medical data to understand generational and regional shifts in political leanings, thereby illuminating methods for slicing and dicing audiences in favor of or against a specified agenda.

Rapidly compiling social media in pursuit of big-data reconnaissance on voters requires artificial intelligence (AI). Machine learning, an AI application where machines learn without explicit programming, will rapidly pore over data troves and illuminate key insights with limited human intervention. Once audiences have been scoped, they’ll need to be prodded, and new innovations will provide scary capabilities for social media audience manipulation.

False information, printed text, spread via news stories true information in the run-up to the election. But fake video and audio can offer strikingly real impressions of world leaders appearing to be in places they’ve never been, saying things they never said. This forgery capability will offer nefarious social media manipulators the ability to inject powerfully engaging smear campaigns into political discussions — or an opportunity to cast doubt about the authenticity of information by alleging that content might be doctored.

Russian interference in Western elections in 2016 heightened concerns about computational propaganda. False social media accounts looking like and communicating like the target audience, known as social bots, repeated programmed messages and amplified political content altering users’ perceptions of reality and influencing debate. 2016’s social bots will appear crude in comparison to the AI-powered chatbots of 2018 and beyond. Newer chatbots, computer programs simulating real conversation, increasingly pass the Turing test, in which a machine exhibits behavior indistinguishable from a human. Bots might seamlessly chat with humans and each other, creating engaging bot communities.

Brad Parscale’s promotion from Trump’s digital director in 2016 to campaign manager for the president in 2020 shows just how important social media campaigns will be in U.S. elections. Campaigns will employ social media not to broaden debate through open discussion, but to harden the views of their social media adherents through deliberate information partitioning. They’ll recruit supporters on mainstream social media platforms and push them to apps they design, control and leverage to harvest voter data.

Over the long term, AI-driven campaigns may well be the undoing of the social media platforms they haunt and the democracies they seek to dominate.

This content was originally published here.

Marc Andreessen famously said that software is eating the world. This notion, that every company must become first and foremost a software company, is hardly a radical notion these days.

However, even as businesses across industries have invested deeply in their software capabilities, they are now struggling with the complexities of modern software development and deployment — software is more distributed, is released in a continuous fashion, and increasingly incorporates aspects of machine learning into the code itself, making the testing and QA function all the more challenging.

Today most enterprise labs require engineers to write testing scripts, and their technical range of skills must be equal to the developers who coded the original app. This additional overhead in quality assurance corresponds with the increasing complexity of the software itself; current methods can only be replaced by systems of increasing intelligence. Logically, AI systems will be increasingly required to test and iterate systems which themselves contain intelligence, in part because the array of input and output possibilities are bewildering.

AI in software testing is already being applied in a variety of ways. Here are three areas in which AI is making the most immediate impact:

Regression Testing

One aspect of testing that is particularly well suited for AI is regression testing, a critical part of the software lifecycle which verifies that previously tested modules continue to function predictably following code modification, serving as a safeguard that no new bugs were introduced during the most recent cycle of enhancements to the app being tested. The concept of regression testing makes it an ideal target of AI and autonomous testing algorithms because it makes use of user assertion data gathered during previous test cycles. By its very nature, regression testing itself potentially generates its own data set for future deep learning applications.

Current AI methods such as classification and clustering algorithms rely on just this type of primarily repetitive data to train models and forecast future outcomes accurately. Here’s how it works. First, a set of known inputs and verified outputs are used to set up features and train the model. Then, a portion of the dataset with known inputs and outputs are reserved for testing the model. This set of known inputs are fed to the algorithm, and the output is checked against the verified outputs to calculate accuracy of the model. If the accuracy reaches a useful threshold, then the model may be used in production.

Machine Vision

Getting computers to visualize their environment is probably the most well-known aspect of how AI is being applied in the real world. While this is most commonly understood in the context of autonomous vehicles, machine vision also has practical applications in the domain of software testing, most notably as it relates to UX and how Web pages are rendered. Determining if web pages have been correctly rendered is essential to website testing. If a layout breaks or if controls render improperly, content can become unreadable and controls can become unusable. Given the enormous range of possible designs, design components, browser variations, dynamic layout changes driven, even highly-trained human testers can be challenged to efficiently and reliably evaluate rendering correctness or recognize when rendering issues impact functionality.

AI-based machine vision is well suited to these types of tasks and can be used to capture a reviewable ‘filmstrip’ of page rendering (so no manual or automated acquisition of screen captures is required). The render is analyzed through a decision tree that segments the page into regions, then invokes a range of visual processing tools to discover, interrogate, and classify page elements.

Intelligent Test Case Generation

Defining software test cases is a foundational aspect of every software development project. However, we don’t know what we don’t know so test cases are typically limited to scenarios that have been seen before. One approach is to provide an autonomous testing solution with a test case written in a natural language and it will autonomously create the test scripts, test cases, and test data.

Among the diverse techniques under exploration today, artificial neural networks show greatest potential for adapting big datasets to regression test plan design. Multi-layered neural networks are now trained with the software application under test, at first using test data which conform to the specification, but as cycles of testing continue, the accrued data expands the test potential. After a number of regression test cycles, the neural network becomes a living simulated model of the application under test.

As AI becomes more deeply embedded in the next generation of software, developers and testers will need to incorporate AI technologies to ensure quality. While it may be a frightening prospect to imagine how a program could train itself to test your apps, it is as inevitable as speech recognition and natural language processing appeared to be a few years ago.

About the Author

Jon Seaton is the Director of Data Science for Functionize, providers of an autonomous software testing platform that incorporate AI and machine learning technologies to automate software development.

Sign up for the free insideBIGDATA .

This content was originally published here.