Metaverse? Quantum!? LaMDa!!?? š¤Æ
Letās start with what is the Metaverse? The word āmetaverseā is a portmanteau of the prefix āmetaā (meaning beyond) and āuniverseā. In my mind, I imagine it as a virtual world we want to live in. The term was coined in 1992 when Neal Stephenson published his hit sci-fi novel Snow Crash, in which the protagonist moves between a virtual world and the real world fighting a computer virus. In the context of Epic Gamesā announcement, the metaverse will be not just a virtual world, but the virtual world ā a digitised version of life where anyone can exist as an avatar or digital human and interact with others. It will be active even when people arenāt logged into it, and would link all previously-existing virtual worlds, like an internet for virtual reality.
Epic Games recently raised 1 billion dollars to build the metaverse. But I imagine the metaverse is so vast that it canāt be build by them alone. So the questions that run through my mind are: Will they create a platform for the world to get involved? Will our work with Augmented Reality(AR)/Virtual Reality (VR) and other immersive technology contribute in making the metaverse a reality? If thats the case and we believe that the metaverse will be a reality in 2030 and beyond, do we as developers and those interested in paving the way for the future need to start considering how we incorporate some of these meta themes into the work we do today? This definitely was eye opening for me and provided some food for thought.
These concepts that make just any metaverse, the metaverse that we want to be part of wouldnāt be possible without some of the work we are doing today. Whether that be Google Quantum computing or LaMDa mentioned at I/O 21.
Language is remarkably nuanced and adaptable. It can be literal or figurative, flowery or plain, inventive or informational. That versatility makes language one of humanityās greatest tools ā and one of computer scienceās most difficult puzzles.
LaMDA, is Googleās latest research breakthrough, adds pieces to one of the most tantalizing sections of that puzzle: conversation.
While conversations tend to revolve around specific topics, their open-ended nature means they can start in one place and end up somewhere completely different. A chat with a friend about a TV show could evolve into a discussion about the country where the show was filmed before settling on a debate about that countryās best regional cuisine.
That meandering quality can quickly stump modern conversational agents (commonly known as chatbots), which tend to follow narrow, pre-defined paths. But LaMDA ā short for āLanguage Model for Dialogue Applicationsā ā can engage in a free-flowing way about a seemingly endless number of topics, an ability we think could unlock more natural ways of interacting with technology and entirely new categories of helpful applications.
There will be many other branches of Emerging technology that will also need to be branched out. But the aspirations and the progress these Companies like Microsoft and Google are making but also the smaller companies is really making the visionary thinking possible
At the Aoteroa AI summit 21, there was a lot of talk about trust, ethics, privacy and responsibility when it comes to AI. These subjects are always topical and seem to always come up. Consider the use of personal data to create a machine learning model in production for your specific use case, now letās say you have customers in the EU where everyone has a right to be forgotten under the General Data Protection Regulation (GDPR). How do we create robust models that are still accurate undergoing ongoing training without the EU customers data? this is a challenge currently being faced by the likes of Microsoft and will be an ever growing challenge that needs to be overcome. In New Zealand we have guidelines primarily intended to be used by New Zealand government agencies/departments and organisations engaging with digital projects involving New Zealand and MÄori Data. This sits under the Treaty of Waitangi/Te Tiriti and MÄori Ethics Guidelines for: AI, Algorithms, Data and IOT which can be found here. It is important for MÄori involvement in the collection and analysis of data. It would be a great injustice creating one set of ethics for MÄori Data or any other system that uses MÄori Data, as MÄori culture is so diverse. If you are confused about this then the Tikanga test will establish which data requires cultural consideration and what aspects of culture can be compromised.
There are still challenges and Risks with AI, including Bias affecting results, Errors causing harm, Data Exposure, Some solutions wonāt work for everyone, and Users needing to trust a complex system. Microsoft outlines from principles of responsible AI such as: Fairness, Reliability & Safety, Privacy & Security, Inclusiveness, Transparency, and Accountability as some of the pillars to help with these challenges. All in all, having these regulations and information about data, and ethics written early is helpful moving forward for all developers and those looking to embark on using AI to make the world a better place.
The narrative around AI can sometimes feel negative, because these headlines are more clickable and sometimes people would rather read about how someone died than how AI has made someoneās life better. Google are doing amazing things with AI for social good. Which is helping provide new ways of approaching problems and meaningfully improve peopleās lives. With AI, we have another tool to explore and address hard, unanswered questions. What if people could predict natural disasters before they happen? Track disease as it spreads, to eliminate it sooner? Or dramatically improve the lives of people with disabilities?
Project Euphoria, stands out to me as I am strong believer in making technology more accessible and itās one of the reasons I think Conversational AI and Natural Language understanding is such a powerful tool for those who have speech disabilities.
This ever changing landscape spans a vast area and makes working in this field extremely rewarding and exciting!
My summary presentation for the 2021 AI summit covered a lot more details into some of these topics. If youād like to see my presentation please find it here.