Technology Trend: The above was written by an AI. The rest is not; it soon could be. This is whats coming The last line of the prologue from the book The Coming Wave by Mustafa Suleyman and Michael Bhaskar gave me no surprise. The whole prologue of a book was written by an AI system answering the following question What does the coming wave of technology mean for humanity? This is what we are headed to.
Since the launch of ChatGPT, Bard-like Generative AI (GenAI) services in the market, AI has come under magnifying glass even more in the whole world. We are standing on the verge of a dilemma, trying to balance like an inverted pendulum. On one hand, we are worried about the threats and risks AI can impose. On the other hand, we don’t want to lose the opportunities AI can bring to industry and society. At this juncture, several institutions and organisations have come up with their predictions in the field of AI in the year 2024. Euronews, PwC and Stanford Human-Centered AI, to name a few, have gone through and prepared a list to be looked upon in this new year. This list caters to international trends and is not specific to any country or region.
Shift in Work Approach and Style
The introduction of GenAI will redefine the work for lower-level employees, middle-level managers, as well as C-suites.
The knowledge base will be highly automated, augmented with GenAI. Most of the operational tasks, like report preparation, financial calculations, tax filings, and legal document preparation, may see a higher degree of GenAI augmentation.
There will be an impact on the consulting business, especially in legal or financial consulting. Organisations that have outsourced these tasks to middlemen or small consulting firms may want to bring them back in house. They will require a small in-house team to operate with the help of GenAI offerings.
Most of the Standard Operating Procedures (SOPs) created to handle different tasks like floor management, inventory management, employee onboarding processes, and Level 1 trouble ticket management will be automated. But everything will not be replaced by AI as humans are required to be accountable as attestants to each and every such activity. But the headcount will decrease, and that work force may shift from one arena to another in the labour industry.
In a nutshell, we will see more implementations of AI which will transform the current transformation happening in industry.
Responsible AI and AI Regulation Will Be In Forefront
There have already been some notions seen since October 2023 in AI regulation. The US came up with Executive Directives. The first international AI Safety Summit took place in the UK where 28 governments, including India, signed a declaration. India hosted the Global Partnership on Artificial Intelligence Summit.Very recently, China issued draft guidelines to standardise the AI industry. This year the whole AI fraternity looks forward to the European Union (EU) passing the first ever AI Act in the world. The EU led the way in data protection and privacy, and regulators across the world expect the same this time as well.
It is inevitable that policymakers will endorse regulations on AI. It is to put guardrails against the ill implementations of AI. We have to contain the inferno of AI misuse in society. These regulations and directives will create some hurdles for the tech industry. Tech giants may try to influence policies and norms. Governments have to be vigilant in the interest of the common public to thwart such attempts. Medium-sized and start-up companies will find it difficult while adhering to all these norms.But eventually, they have to comply to create a better ecosystem.
This year we shall see more organisations rolling out their AI data protection frameworks to handle client and internal data just as they did for digital data privacy and protection.
AI Safety Institutes are to be constituted in the USA and the UK. They may start functioning this year. These institutes will have early access to various AI products and services for analysis and review purposes before they are rolled out in the market.
There will be a political tussle among countries to lead the AI regulation initiatives. But we need more and more international collaborations to walk in this path.
Deepfake Proliferation
This is one of the cons of AI, or, to be specific, GenAI. This needs a special mention, as deepfake has the ability to create instability in society. Well see many more videos in the coming days where actual people will not do certain activities or utter certain narratives, whereas their deepfake videos will.
Deepfake videos are created with GeneAI, where someones personality can be morphed and fake videos can be created. Researchers are very worried and specifically mentioned the damage deepfake can create during the time of election. It is alarming for India as well, where parliamentary elections will be held in the coming months. Political parties may use such videos to influence vote banks.
GPU Shortage
AI models need huge computational power. A special type of processor is required to withstand this computation. They are called Graphical Processing Unit (GPU). Now, in majority of the cases, these AI models are accessed through the cloud providers which provide very fast computing. But still at hardware level at data centers, these units require more operating capability using less electrical power.
With the advancement of AI, the world will feel the urge of more powerful GPUs to run those AI models in 2024. But the chip manufacturers may be low in stock. Only a few big chip manufacturers like NVIDIA can withstand that overshooting demand. Stanford University is extensively doing research to find low power alternatives to current GPUs.
AI Will Give Birth To New Products And Services
AI will bring new types of services to the industry. I shall mention two such potential areas with examples below.
The PwC 2024 AI Business Predictions report says that GenAI is going to be the missing link for data in 2024. Why?
GenAI has the capability to find insights from unstructured data available in your organisation for your client. Gen AI can transform data into information very quickly. Your offline physical data (text, image or video) can be scanned by GenAI systems; a summary or outline can be created for better labelling; and only the relevant copies can be retained in your organizations secure data vault. All these can be achieved at a rapid pace. Thus, organisations can declutter their knowledge data repository.
We already interact with chatbots on many websites to raise our complaints, know our status, to get help for certain activities ( example- on banking websites, we ask how to open a savings account and chatbot gives the links and necessary pointers). This type of service will be enhanced multifold with the power of AI. More helpful chatbots will be introduced. Say for example – You ask your Siri or Alexa or Google Assistant to book a flight ticket. That will connect to an AI agent on a flight ticket booking site and share your requirements and details.That AI agent on the booking site will book a ticket for you and send it back to your mail id. Cross-platform communication between AI agents will start to flourish in the days to come. Remember the video shown at Google I/O event in 2018 where a Google Duplex AI agent made a call, had a human-like conversation with a human saloon staff to book an appointment? This time, it will be one level up. A bot will be interacting with another bot.
The Dawn Of Emotional AI And AGI
This year, we can see some breakthroughs in the field of emotional AI. It is a specific use case where AI systems can capture your emotions from your facial expressions through video and make a decision. Lets assume that you are watching a movie on an AI-integrated streaming platform on your phone. You have enabled the camera to record your video while watching. By analysing your facial expressions, the platform can show a message to switch to another movie if you’re not enjoying the same. Also, it will less recommend movies falling in that same category. It will do the reverse in case you enjoy it.
The discussion will revolve around AGI – Artificial General Intelligence, where AI systems can perform cognitive tasks just like humans. It can think about how we think, how we experience, receive, process information, and remember things. But this has paramount complexity within it and is at a very early stage.
Ten years back people used to say that AI could only automate repetitive tasks.The introduction of GenAI systems has shown us that it can perform creative works to a certain degree. Advancement in AI will definitely open many new avenues but our focus should be on augmenting people with AI systems, not replacing them. We have to safeguard those systems to restrict misapplication. The time has come to be more Homotechnologicus (A term coined by Mustafa Suleyman in his book, indicating technological animals) than ever for all of us through proper education, skill development and awareness. In every step of this giant leap in technological enhancement, we should be very clear about what we want from AI.
By Anjuman Bhattacharyya, Solution Architect, Wipro Ltd
