Machine learning and AI are being used more often on the front lines of the cybersecurity battle. They can, among other things, help identify threats more efficiently, and also significantly improve response times. But the “bad guys” also benefit from machine learning and AI. For example, machine learning can produce attack scripts at a speed and level of complexity that humans can’t match. AI can also help hackers avoid security detection, for example by using AI to speed up malware, causing it to constantly change its code, and therefore allowing them to bypass many security controls. This expert panel will explore the best, and the worst, cybersecurity use cases for ML and AI, and discuss the preventative measures that can be taken to avoid giving the hackers the upper hand in the data security wars.

A trademark identifies and distinguishes the source of the goods of one party from those of others. But will this source identifying function still be relevant in a world where AI can remove humans from much of the product selection and purchasing process? When making everyday purchases for us, will Alexa, Google Assistant, or Cortana look to identify specific brands? Or will they just look to accurately and efficiently match its owners’ needs with any products that fulfill those needs? This future state could essentially limit the effectiveness of the brand as an identifier, and would require us to rethink many of the classic principles of trademark law (for example, likelihood of “consumer” confusion). Some principles would simply no longer apply, while others would have to be reinterpreted to reflect the new paradigm. We will explore this issue in depth, as we look to understand the changes that will be necessary to prepare for the near future.

With the uncertainty of Congress being able to pass federal privacy legislation, states are leading the way in enacting robust privacy laws to protect consumer privacy. Join a panel of attorneys from the state AGs as they discuss and answer questions relating to the latest issues surrounding state privacy regulation, enforcement trends, and a look into the future of state privacy regulation.

Join your colleagues for a review of the most noteworthy copyright cases of the past year!

Protecting data in an Internet of Things environment will be extremely difficult. The essence of IoT is about making connections, and using those “connected” things to gather information, send information, or both. But all of this data collection and sharing creates new opportunities for information to be compromised. An example being, a Deloitte article raised a question about a garage door opener that could also deactivate a home alarm upon entry, such that the entire alarm system could potentially be deactivated when only the garage door opener is compromised. Equally as troubling, in 2017, the US FDA issued a statement warning that certain implantable cardiac devices, which connected wirelessly to the internet, could potentially be accessed by malicious hackers. Connection is the whole point of IoT, but we need to make sure the connections do not allow malicious actors to gain entry into IoT ecosystems, access customer information, or penetrate manufacturers’ back-end systems to launch malicious attacks. Our experts will discuss in detail what needs to be done to build the IoT infrastructure safely, so as to protect the integrity of IoT innovations and ecosystems.

Join your colleagues for a review of the most noteworthy trademark and unfair competition cases of the past year!

The California Consumer Privacy Act (CCPA) was enacted in June 2018, and takes effect in January 2020. The Act requires companies doing business in California to improve transparency in how they collect, use and share data. The law is not a copy of the EU’s GDPR legislation, but there are similarities (particularly in the way they both define “personal information” and information “processing”). Privacy laws are also starting to appear around the US, with Vermont, New Jersey, Washington State and others considering legislation. And, while Congress has not actually proposed federal privacy legislation, those reading the tea leaves are starting to whisper that it’s only a matter of time. While there are strong views on both sides of this debate (and already a battle brewing over whether the CCPA would be preempted by a federal law), a recent committee hearing on this topic seemed to suggest that there is at least an acknowledgement by many that technology has outpaced the law, and that one federal privacy law could potentially be better than the collection of 50 different state laws. Join our panel, and help identify the best path forward, as we look to create the American approach to data privacy.

Data security experts are extremely concerned about both current threats and future trends. Whether it’s the aggressive move of data to the cloud (often without adequate planning and/or adequate funding), the unwillingness of some companies to leverage the lessons learned by others, the use of new tools such as collaboration apps that can increase exposure, or the constant stream of new attack vectors used by criminals such as formjacking and powershell attacks, corporate victims often feel like they are playing the classic the carnival game “Whack-a-Mole.” Our expert panelists will discuss the current threats and trends, and what the good guys should be doing to protect themselves, their businesses, and their customers.

Warren Buffet once said: “It takes 20 years to build a reputation and five minutes to ruin it. If you think about that, you’ll do things differently.” Applying this truism to data security, and the profound damage that a data breach could have on a brand’s reputation, leaves many asking what we should be doing differently. The answer is, quite a bit actually. Our data security experts will address issues from the most basic policies and controls to the more complex strategic cyber security postures. They will also provide roadmaps for creating and maintaining sound data security practices for small companies, large companies, governments, and law firms of all sizes.

Join your colleagues for a review of the most noteworthy data security and privacy cases of the past year!

While this is not the first time that IP practitioners and judges had to wrestle with the concept of authorship, the AI use case is likely to present the most difficult challenge yet. Can AI generated works be protected by copyright? If so, who is the author? The majority of experts today take the position that only humans can be authors under copyright law, and would consider a work authorless or in public domain if there was insufficient involvement of a human creator. Several questions remain, however. First, as a community of creators and protectors, how can we help define when a human has contributed enough (either as a programmer or a user) to be an author of a work. Second, are there, and should there be, other ways to protect such innovations, such as sui generis or considering AI generated outputs as “works made for hire” by an “employed” AI. Third, would any of the above change if, in the future, a truly self-aware, autonomous AI submitted a copyright work for registration? Our experts will examine all of these questions and more.

The focus on privacy issues is a growing global phenomenon. China, for example, introduced an information security specification in May of 2018 (and updated it in February 2019), which borrows quite a bit from the GDPR in terms of substance. It’s a nonbinding National Standard, but some have noted that authorities are referring to it when enforcing more generally worded provisions found in mandatory Chinese laws. In India, the Supreme Court in 2017 affirmed that privacy is a fundamental right, based on the rights of life and liberty in the Constitution. And in 2018, India introduced a draft data protection bill that borrows heavily from a number of the concepts in the GDPR. Under the Consumer Data Right (CDR) in Australia, citizens will be able to “own” their personal data through more open access to banking information, energy details, phone and internet business. They will also have more control regarding the use of their data and who can access it. However, for companies that do business globally, one of the most concerning issues is not about how to comply with these new laws in isolation, but rather how to manage the complexity of this patchwork of global privacy standards. Join a spirited debate, led by professionals from the US and abroad, about the various approaches to privacy around the world, and best practices for managing global privacy obligations properly and efficiently.

In the past, instances of infringement or unfair competition would almost always result in the sending of a quintessential, strongly worded, cease and desist letter. In the age of social media, however, when your C&D can be posted and heavily criticized on multiple social media platforms, many companies and their lawyers have opted to take a different approach. Perhaps a toned-down version of a classic C&D is a good fit. Or maybe a touch of humor would work best. Velcro famously opted to have singing lawyers verbally deliver a C&D, while Anheuser-Busch hired an actor dressed as a town crier to read a C&D aloud (both of which generated positive press). In some cases, a simple phone call is best. There is also the option of taking no action at all in cases that involve a minor or sporadic unauthorized use of a mark. Don’t miss this opportunity to engage with your peers and learn how to master this new form of art!

There is a titanic battle taking shape between system defenders and hackers. And some industry veterans are fearful that system defenders are jumping on board the AI/ML bandwagon without paying enough attention to risks, such as relying on a single master algorithm or using algorithms with “explainability” issues that make it hard to assess what’s driving anomalies. Others worry more about the tendency to go it alone. According to Jason Matheny, Director of Intelligence Advanced Research Projects Activity at the Office of the Director of National Intelligence (DNI), the advantage boils down to which side has the most training data, and that the arms race could be won by the “good guys” if we worked together. For example, if the large technology companies pooled all the data it had on cyberattacks, would we have solved the problem because that data set would be greater than that of any attacker? Or is the solution something else entirely? Join us as we explore areas of consensus, and ways to work more closely together to counter the threat of malicious cyber attacks in the future.

AI training, which is a fundamental part of any AI project, has been in the news of late. For example, a Google AI was trained to speak more like a human by being fed human-created novels. A Sony/Flow Machines AI was trained to create music by being fed human-created songs. And an IBM AI was trained to recognize faces by being fed human-created photographs taken from Flickr. Even assuming that AI tools need to be trained this way for now (until it can emulate the more top down reasoning that more closely resembles the way that humans approach problems), there are still questions about what is, and what is not, fair. Should companies that use copyrighted material to train AI be required to ask permission first, and/or to compensate the underlying creators? Or is the AI training activity a fair use? If it’s not fair use, how do parties that want to “consume” creative works in this fashion go about seeking consent from hundreds or even millions of creators? What about larger societal issues? Some have called for changes in law and policy to promote social equity in the AI age, such as universal basic income and “robot taxes.” These are all questions that still need to be answered. Don’t miss this fascinating discussion about copyright’s place in the Fourth Industrial Revolution.

Human influencers have partnered with every brand from Amazon to MeUndies. When they do, the Federal Trade Commission’s Endorsement Guides dictate how they disclose the partnership, usually with hashtags like #ad or #sponsored. More recently, however, there is a trend toward using fictional CGI influencers. And followers are often not aware that these influencers are not human. Lil Miquela and Shudu, for example, are popular CGI influencers that only recently revealed that they are not human. This trend raises a lot of unanswered questions: Should their creators be required to disclose that they are not human? Do they use the same disclosure requirements as human influencers? Are there ethical issues with realistic digital characters promoting an unattainable representation of beauty? Should there be a new category of disclosure statements, such as #RealPerson? Our panelists will walk us through the provocative legal and ethical issues created by these new brand ambassadors.

Better IP understanding is not just for lawyers and rights owners. Creating an environment where inventions, content, and brands are respected means higher valuations and better ROI. Establishing a basic understanding of the purpose of IP rights and how they impact people, commerce and jobs are more relevant than ever. Misinformation impacts shareholders and consumers alike.

Description coming soon

Description coming soon

In 2017, Facebook’s Artificial Intelligence Research team was reportedly doing experiments with dialog agents (AI systems designed to communicate with humans), when the chatbots started communicating with each other using a language that they made up. In 2018, Walmart filed a patent for a shopping cart that monitors various biometric signals, like a customer’s body temperature, heart rate and tightness of grip on the handle. The data would be transmitted to a server and used to alert staff when shoppers might need help or be dissatisfied. Also in 2018, researchers at MIT trained its “Norman” AI on image captions taken from a community on Reddit that is notorious for sharing graphic depictions of death. After “training” that way, Norman was given a Rorschach inkblot test. Looking at an image that a “normal” AI identified as a black and white photo of a baseball glove, Norman saw a man being murdered by machine gun in broad daylight. These examples (and many others) raise a question about whether, with AI and other groundbreaking innovations, we should not only ask “can we” but should also consider “should we.” And when the answer to both is yes, how do we determine when additional precautions and/or protocols should be followed along the way? Join us as we discuss and debate the meaning of ethical innovation.

The session will focus on the evolving nature of intellectual property in China and its role in enabling innovation in what has become an increasingly software-driven and open source-centric world.

A.I. enables new mass disinformation techniques including Deep Fakes, Artificial online profile pictures, and profiles. Social Networks have the tools to block them, but not the will to do so.

Description coming soon

Algorithms are a crucial building block in the Fourth Industrial Revolution. But there are a number of ethical issues arising from their use. For example, in a recent study researchers ran ads on a social media platform for several different types of jobs. The ads were delivered as follows: Ads for janitors were delivered to an audience that was 65 percent female and 75 percent black; ads for jobs in the lumber industry were sent to users that were 90 percent male and 70 percent white; ads for supermarket clerks were shown to an audience that was 85 percent women, and ads for taxi driving opportunities were delivered to audiences that were 75 percent black. Overall, the experiment suggested that even when businesses affirmatively choose to show their ads to wide audiences, the ads may still get sent to specific recipients by the advertising platform based on race or gender stereotypes. Identifying, and then correcting, these and other algorithmic issues will be a gargantuan task, but it can be done. Our experts in business, law and government will explore this critical issue in depth.

Description coming soon

In 2000, Pamela Samuelson published an article on internet privacy titled “Privacy As Intellectual Property?” If you were to read that article again today, you might wonder what if anything changed in 19 years. For example, the article notes: “The technical infrastructure of cyberspace makes it remarkably easy and cheap to collect substantial amounts of information identifiable to particular individuals.”“Once these data have been collected, information technologies make it very easy and cheap to process the data in any number of ways (for example, to make profiles of particular users’ interests).”“The market incentives for firms to collect and process personal data are very high. Data about users is not only useful in assessing how a firm might improve its service for its customers, but it also has become a key commercial asset which firms use both for internal marketing purposes and for licensing to third parties.” While Professor Samuelson’s article endorses a licensing approach to the protection of personal data, it recognizes that the law alone cannot solve information privacy problems in cyberspace. She writes that: “Work must continue on evolving norms about appropriate and inappropriate uses of personal data, on persuading firms that the trust necessary for electronic commerce to flourish requires the interests of individuals in information privacy to be given appropriate deference, and on adapting the technological infrastructure of cyberspace so that information privacy becomes easier to achieve.” Join us as we go back to the future to re-examine the issue of data privacy online.

Description coming soon

Description coming soon