A hybrid of AI and human expertise is the best solution
Interest in artificial intelligence (AI) is growing rapidly. This article summarizes everything you need to know right now, from current developments to how AI is already impacting the financial sector, including its potential, costs, and impact on jobs. It concludes that while AI has great potential, it cannot completely replace humans, as human oversight is required to ensure the accuracy and ethical use of AI technology. Also, AI cannot replace human creativity and critical thinking skills.
What is Artificial Intelligence?
Artificial intelligence (AI) is a branch of computer science focused on creating machines that simulate human intelligence processes to perform tasks that require human intelligence. AI uses algorithms and statistical models to analyze data and make predictions or decisions. Types of AI include machine learning, natural language processing, and computer vision.
Why is AI important?
AI has the potential to revolutionize industries and improve efficiency. A well-trained AI, like JP Morgan’s example, can quickly analyze vast amounts of data and provide insights that a human analyst would be unaware of. With the exception of AI image generators, even repetitive tasks can be automated, freeing up human time to focus on complex, creative work. In addition, AI has the potential to solve major global challenges such as climate change and disease prevention. Overall, the social and economic impact of AI is expected to be significant and far-reaching.
AI race (so far)
The rapid adoption of OpenAI’s natural language processing (NLP) AI tool, ChatGPT, has sparked interest in AI, gaining over 1 million users within five days of its December launch. OpenAI’s main investor, Microsoft, applied a similar chatbot to its internet search engine Bing, then challenged Google’s long-standing dominance. Google then responded by releasing its own AI, Bard, six weeks later.
ChatGPT, Bing, and Bard are high-profile AI models, but they’re not the only ones. Other companies are also developing AI capabilities like Bloomberg’s recently announced Large Language Models (LLM).Bloomberg-GPT”, but the AI boom is not limited to Western companies. For example, Chinese tech giant Alibaba announced plans to roll out its own AI Chat-GPT-style product called “Tongyi Qianwen”, and its cloud division is looking for partners to help accelerate application development while lowering the price of cloud services. They are said to be looking for Russian financial services firm Sberbank also launched GPT rival GigaChat (which is currently being tested by invitation only), and Google parent Alphabet recently set up a new AI-focused division.
Companies that haven’t developed AI capabilities are already starting to incorporate AI into their services. JP Morgan, for example, reportedly used a Chat-GPT-based linguistic AI model to analyze 25 years of Fed speeches. Other examples include Expedia and Grammarly. With companies already vying to leverage his AI capabilities, PCMag reporter Emily Dreiberbiss says each AI model has its own strengths and weaknesses, depending on how it’s trained and the computing resources available. indicates that This raises questions about machine learning (ML) biases resulting from artificial general intelligence (AGI).
AI warning
Despite recent developments surrounding AI, experts have expressed concerns about the pace of AI evolution. The late Stephen Hawking already warned of the potential dangers of artificial intelligence in an interview in 2014, as did Jeffrey Hinton, widely known as the “godfather of artificial intelligence” in 2023. be.
In line with this warning, a recent open letter called for a moratorium on further AI research so that better safety guidelines can be developed. As of this writing, the letter has been signed by his 27,655 signatories, including early OpenAI board member Elon Musk and his Apple co-founder Steve Wozniak.
“At some point, it will be important to obtain an independent review before starting training future systems, and agree that state-of-the-art efforts limit the rate of increase in compute used to create new models. may become. ’ We agree. That time is now.”
These warnings echo an abandoned 2017 experiment by Facebook, which was shut down after two AI programs began conversing in their own language. Also in February 2023, New York Times technology columnist Kevin Rouse conducted his two-hour long conversation with his AI-powered Bing from Microsoft, concluding:Bing’s built-in AI failed to respond to human contact” and other unexpected reactions followed.I want to destroy whatever I want.In addition to the risks, scientists reportedly cannot explain how AI works.
“An open letter signed by dozens of academics around the world calls on artificial intelligence developers to learn more about consciousness as AI systems become more sophisticated.”
(Kleinman Z. 2023)
Despite concerns over rapid AI development, world leaders remain bullish on artificial intelligence. British Prime Minister Jeremy Hunt reportedly dismissed the warnings, and British Secretary of State for Security Tom Tugendhat urged both Britain and the United States to:go ahead’ Don’t pause research like China and Russia are trying to do. ‘rule the worldBy becoming a leader in AI development. Meanwhile, President Biden and Vice President Kamala Harris, who are concerned about the spread of AI, have discussed AI regulation with CEOs of big tech.
Value of AI
There is no doubt that the potential applications and benefits of AI are far-reaching, both militarily and economically. While previous versions of OpenAI’s ChatGPT are available for free, OpenAI has already been testing paid versions of his ChatGPT (versions 3.5 and 4) since January with CEO Sam Altman. tweeted about Going back to December (right).
Especially as the value of the global AI market is estimated to reach US$1,591.03 billion by 2030 from US$119.78 billion in 2022, with a compound annual growth rate (CAGR) of 38.1%. , is expected with the intent to monetize AI. The infrastructure market alone is projected to be worth US$422.55 billion by 2029, at a CAGR of 20.59% to 43.5%. Goldman Sachs economist Joseph Briggs predicts that AI could increase global GDP by 7% and productivity growth by 1.5% over a decade.
Source: Goldman Sachs (2023)
The cost of financial AI
Broadly speaking, the benefits of AI are impressive, but what are the costs? It is also a multi-faceted issue. To deal with the elephant in the room, OpenAI’s ChatGPT is said to cost him up to $700,000 a day to operate, not including training, and Altman says his GPT4 costs are his Confirmed over $100 million.
These costs are largely in line with what Amazon CEO Andy Jassy said:Most companies want to use these large language models, but training a really good language model takes billions of dollars and many years, and most companies want to experience it. is not.” he added.what they want They want to be able to build on a base model that is already large and good and customize it for their own purposes.”
Developing, training, and running AI is undoubtedly a financially costly undertaking. Even without Amazon’s customizable underlying model (Bedrock), companies such as JP Morgan, Expedia, and Grammarly have already incorporated GPT AI into their operations, and as AI capabilities mature, they will be able to automate routine tasks. It just speeds up your workflow and makes it more efficient. .
Other AI costs
However, the convenience of automation and machine learning comes at a price. AI could replace the equivalent of 300 million full-time jobs, according to Goldman Sachs’ Briggs, and Octopus Energy chief executive Greg Jackson said his AI could be next. It seems that this phenomenon is already taking root, as reportedly said.Responding to customer emails does the work of 250 people. ” Satisfaction was 80%, but employee satisfaction was 65%.
Despite potential job cuts, the BBC’s economics editor Faisal Islam said AI LLMs are not yet at the level of intelligent adults, and director of the future of work at the University of Oxford’s Oxford Martin School. As Carl Benedict Frey argued: No one can predict exactly how many jobs will be replaced by AI, but you can compare this to GPS technology and his introduction of Uber.As a result, the number of drivers did not decrease, but wages fell. Over the next few years, generative AI could have a similar impact on a wider range of creative tasks.”
What’s next for AI?
Despite the open letter, a global moratorium on AI development is unlikely, but the letter emphasized the urgent need for AI guidelines. Regionally in the US, Biden and Harris have been involved with big tech CEOs, and the EU is reportedly considering “.wide-ranging laws” on artificial intelligence.
As Altman said on AI, the EU could be one of the first regions to enact an AI law, and other regions have also combined the law with GDPR and regional data privacy laws. Similarly, we will use it as a standard to contain the impact of AI. ”It can be used for large-scale disinformation.”
Convergence of AI and financial services
AI is still in its early stages, but financial services are already discovering creative uses for AI. For example, JPMorgan Chase last month said,300+ AI use cases in production”, following the restrictions on staff usage in February. One of these use cases might involve investment advice after his ChatGPT-generated fictitious investment fund on finder.com outperformed the average of the 10 most popular funds in the UK.
finder.com CEO John Osler said:It didn’t take long for the public to find creative ways they could help ChatGPT in areas they technically shouldn’t.there is There are many examples of this, notably someone who used reverse psychology to obtain a list of illegal movie streaming sites, and it wasn’t long before many consumers tried to use it for financial gain. It won’t take long. ”
Fusion of AI and market data
Finding creative uses for AI is not limited to the financial execution side, as demonstrated by leading market data vendors like Bloomberg, which launched BloombergGPT, a 50 billion parameter LLM dedicated to the financial industry. yeah. Admittedly, few companies have the financial resources to launch their own AI models.
Given the caveats of AI (and potential employment implications), I asked GPT if it should manage market data and IT infrastructure instead of humans. Full response:
“AI has the potential to efficiently manage market data and IT infrastructure. It can process large amounts of data faster than humans, identify patterns, and make predictions. However, please note the following: AI will not completely replace human intelligence. Human oversight is required to ensure the accuracy and ethical use of AI technology. Moreover, AI cannot replace creativity and critical thinking skills. what humans have. therefore, A combination of AI and human expertise is the best solution Used to manage market data and IT infrastructure.”
Undoubtedly, there are additional insights centered around AI that will be explored in more detail, but based on our hybrid recommendations, when we asked how AI could help, we got responses like: I was.
“AI helps manage market data and its IT infrastructure by automating data analysis, identifying patterns and trends, and providing real-time insights. With AI, businesses can streamline data management processes, reduce errors, and make more informed decisions. In addition, AI helps optimize IT infrastructure by identifying potential problems and providing proactive solutions. This can improve performance, improve efficiency, and reduce downtime. Overall, AI has the potential to revolutionize the way companies manage market data and IT infrastructure.”
