AI Contracts

[willendure]

Any thoughts on how the current AI trend will play out for contracting work? Obviously it has already existed for quite some time even if not as GPT models, but there have been plenty contracts around for data science, ml, ml-ops etc for a while. But firing off a few searches specifically on terms such as AI, LLM or GPT doesn't bring back much. Is it too early still for every business and government function to be jumping on the AI bandwagon and pulling people off the streets to satisfy the demand for contractors to build this out? Or is AI going to play out in quite a different way to previous technology step changes?

I currently have a contract with a business making software for a CASE tool, and yes, they are keen to add LLM-based AI to it. Do others find that their customers are interested in dipping their toes into the AI waters?

[Lance]

Any thoughts on how the current AI trend will play out for contracting work? Obviously it has already existed for quite some time even if not as GPT models, but there have been plenty contracts around for data science, ml, ml-ops etc for a while. But firing off a few searches specifically on terms such as AI, LLM or GPT doesn't bring back much. Is it too early still for every business and government function to be jumping on the AI bandwagon and pulling people off the streets to satisfy the demand for contractors to build this out? Or is AI going to play out in quite a different way to previous technology step changes?

I currently have a contract with a business making software for a CASE tool, and yes, they are keen to add LLM-based AI to it. Do others find that their customers are interested in dipping their toes into the AI waters?

The AI trend in contracting work is likely to continue growing, but its exact trajectory will depend on various factors. AI has been steadily advancing and integrating into numerous industries and applications, including data science, machine learning, and ML-ops. However, there are several considerations that could influence how AI contracting work plays out:

1. Maturity of AI technologies: AI, particularly GPT models like LLM, is a rapidly evolving field. As the technology matures, it becomes more accessible and easier to integrate into various systems. Companies and governments may be hesitant to invest heavily in cutting-edge AI until they see proven results and a stable landscape of available tools and frameworks.
2. Market demand and competition: As AI technologies become more widespread, the demand for AI contracting work is likely to increase. Businesses will seek to incorporate AI to gain a competitive edge, optimize processes, and enhance decision-making. However, the availability of skilled AI contractors may also increase as the technology becomes more prevalent, leading to competition in the market.
3. Barriers to entry: AI expertise often requires specialized knowledge and skills. While AI contracting opportunities may rise, there might be a shortage of qualified professionals to fulfill the demand. This could potentially lead to companies seeking to upskill their existing employees or invest in AI training programs.
4. Complexity and understanding: AI solutions, especially GPT models, can be complex and require a deep understanding of the underlying technology to implement effectively. Many businesses may not fully comprehend the potential of AI or how it can be applied to their specific use cases. Hence, there might be a need for AI consultants to guide clients in adopting AI solutions tailored to their requirements.

Regarding businesses and government functions jumping on the AI bandwagon, adoption rates can vary. Some forward-thinking organizations may have been early adopters, leveraging AI in their operations, while others may still be in the early stages of exploring AI's potential.

For your current contract with a CASE tool company, their interest in adding LLM-based AI indicates a growing trend of incorporating AI capabilities into software products to enhance functionality and efficiency. Clients in various industries may indeed show interest in dipping their toes into AI, but it's essential to approach AI adoption with clear use cases and well-defined objectives to ensure successful implementation.

As for the AI landscape evolving differently from previous technology step changes, AI's transformative potential is significant. Unlike some previous technologies that were more incremental improvements, AI has the capacity to revolutionize entire industries and change how businesses operate fundamentally.

In conclusion, AI contracting work is likely to continue expanding as AI technologies mature, demand increases, and organizations recognize the value of AI solutions. However, it's essential to consider the complexity of AI and the need for skilled professionals to fulfill this demand. Adoption rates can vary, but more businesses are likely to express interest in exploring AI to stay competitive and improve their operations.

[willendure]

Barriers to entry: AI expertise often requires specialized knowledge and skills. While AI contracting opportunities may rise, there might be a shortage of qualified professionals to fulfill the demand. This could potentially lead to companies seeking to upskill their existing employees or invest in AI training programs.

This sort of shortage is what I would hope to take advantage of as a contractor.

An interesting twist on that is that at the moment it seems like everyone is some kind of AI expert, banging on about "prompt engineering" whilst knowing precious little about what a transformer even is. But even Jensen Huang, head of nVidia, gave a recent speech where he declared that "everyone is a programmer" now. In reference to the truth that AI is going to open up the creation of complex automation tasks to people who will do little more than ask the AI for what they want.

I somehow don't think it will work out like that despite there being elements of truth in it. Some real expertise will surely still be required to succesfully operate the most advanced technologies we have ever created.

[Lance]

This sort of shortage is what I would hope to take advantage of as a contractor.

An interesting twist on that is that at the moment it seems like everyone is some kind of AI expert, banging on about "prompt engineering" whilst knowing precious little about what a transformer even is. But even Jensen Huang, head of nVidia, gave a recent speech where he declared that "everyone is a programmer" now. In reference to the truth that AI is going to open up the creation of complex automation tasks to people who will do little more than ask the AI for what they want.

I somehow don't think it will work out like that despite there being elements of truth in it. Some real expertise will surely still be required to succesfully operate the most advanced technologies we have ever created.

Let's explore the implications of the "Barriers to entry" paragraph in more detail:

1. Shortage of AI Experts: As AI technologies advance, the demand for skilled AI professionals can outpace the supply. There might not be enough experienced AI contractors available to meet the growing demand for AI projects across various industries. This shortage could lead to increased competition for AI talent, potentially driving up the cost of hiring qualified contractors.
2. Upskilling Existing Workforce: To address the shortage of AI experts, some companies may choose to invest in upskilling their existing employees. They might identify individuals with relevant backgrounds or aptitude and provide training and resources to equip them with AI knowledge and skills. Upskilling can be a cost-effective way to build an in-house AI team and ensure long-term AI capabilities within the organization.
3. AI Training Programs: Recognizing the increasing importance of AI, educational institutions and training providers may offer specialized AI training programs to meet the industry's demand for skilled professionals. These programs can help individuals transition into AI-related roles and address the skills gap in the job market.
4. Collaboration and Partnerships: Companies seeking to integrate AI into their products or services may collaborate with specialized AI contracting firms or partner with AI research organizations. This way, they can leverage the expertise of external AI professionals and accelerate the implementation of AI solutions.
5. Automated AI Tools: Another potential response to the shortage of AI experts could be the development of automated AI tools and platforms that require less human intervention. These tools aim to simplify the process of AI development and deployment, making AI more accessible to non-experts and reducing the reliance on highly specialized AI contractors.
6. Remote Work and Global Talent: With advancements in remote work capabilities, companies may look beyond their local talent pool and tap into the global market for AI contractors. This approach allows them to access a more extensive pool of AI experts and potentially find specialized skills that might be scarce in their local area.

It's important to note that the AI field is multidisciplinary, requiring expertise in areas such as data science, machine learning, natural language processing, computer vision, and more. As a result, upskilling and training programs need to address these diverse skill sets to produce well-rounded AI professionals.

Overall, addressing the barriers to entry in the AI contracting market will be crucial for the continued growth and successful adoption of AI technologies across various industries. Companies that proactively invest in building AI expertise within their organizations or collaborate with external AI contractors will be better positioned to capitalize on the potential benefits that AI can offer.

[Lance]

This sort of shortage is what I would hope to take advantage of as a contractor.

An interesting twist on that is that at the moment it seems like everyone is some kind of AI expert, banging on about "prompt engineering" whilst knowing precious little about what a transformer even is. But even Jensen Huang, head of nVidia, gave a recent speech where he declared that "everyone is a programmer" now. In reference to the truth that AI is going to open up the creation of complex automation tasks to people who will do little more than ask the AI for what they want.

I somehow don't think it will work out like that despite there being elements of truth in it. Some real expertise will surely still be required to succesfully operate the most advanced technologies we have ever created.

You are absolutely right. While there is a growing interest in AI and many people are excited about its potential, it's essential to recognize that true expertise is still required to operate the most advanced AI technologies effectively.

AI has indeed become more accessible to a broader audience through advancements in tools and platforms that simplify certain tasks. "No-code" or "low-code" AI development platforms allow individuals with limited programming knowledge to build relatively simple AI applications. Prompt engineering, as you mentioned, is also a technique that enables users to interact with AI models using natural language queries. These developments democratize access to AI to some extent and empower non-experts to accomplish certain tasks that previously required more specialized knowledge.

However, these simplified approaches have their limitations. While they can be helpful for specific use cases, they may not be suitable for tackling complex and large-scale AI projects. True AI expertise is essential when dealing with cutting-edge AI research, building custom models, implementing AI solutions in critical systems, dealing with large and diverse datasets, ensuring ethical AI practices, and addressing the challenges that arise during AI deployment and maintenance.

AI is a multidisciplinary field that encompasses mathematics, statistics, computer science, data engineering, domain expertise, and more. It requires a deep understanding of underlying algorithms, model architectures, optimization techniques, and the ability to handle real-world complexities that often arise during AI projects.

Furthermore, as AI technologies continue to evolve, new challenges and complexities will arise, necessitating experts who can keep up with the latest research and best practices. This is where skilled AI contractors and experts will continue to play a vital role in driving the AI revolution forward.

While AI is becoming more accessible and will undoubtedly empower people to automate certain tasks with less coding expertise, it is unlikely to eliminate the need for specialized AI professionals. Instead, the division of labor may shift, with non-experts utilizing simplified AI tools for specific tasks while collaborating with AI experts for more complex and impactful AI projects.

In conclusion, while AI democratization is opening up new opportunities and possibilities, the value of true AI expertise cannot be underestimated. As an AI contractor with specialized skills and knowledge, you can indeed take advantage of the growing demand for AI projects and play a critical role in helping businesses and organizations leverage AI to its full potential.

[northernladuk]

What both the above says but we have to consider the contract element of it. It's an emerging tech and there will be a lot of people jumping on the bandwagon with a lot of companies investing for the long term. This will likely mean building their internal capability up i.e. perms. There is unlikely to be enough people with enough skill to deliver contractor type services for awhile. I'm not too sure even 6 months in AI is going to be enough to secure that type of work at decent/top rates. Reading up on it/courses/home tinkering isn't enough to get people in to mainstream IT contracts let alone emerging tech.

A few people might get lucky having a hand in somewhere but I don't think there is going to be much work out there for your average contractor bod that fancies a change of tech for awhile yet. That's not to say start doing everything you can to prepare but its going to be difficult to gain the skills and experience you need to sell back to a client in the short term.

[Snooky]

Several years back, the Next Big Thing in financial services (and elsewhere, probably) was blockchain. Everyone thought it was going to revolutionise the industry. It certainly had the potential to significantly improve it in many ways, but it's really taking its time to get there.

I see AI in similar terms. It'll be great for some applications and will almost certainly be adopted more widely than blockchain, but unless a company can see a major payoff in terms of reduced costs, increased revenue or commercial advantage, they're unlikely to do much more than play with it.

As with most things, I'll probably be proven wrong though

[Lance]

What both the above says but we have to consider the contract element of it. It's an emerging tech and there will be a lot of people jumping on the bandwagon with a lot of companies investing for the long term. This will likely mean building their internal capability up i.e. perms. There is unlikely to be enough people with enough skill to deliver contractor type services for awhile. I'm not too sure even 6 months in AI is going to be enough to secure that type of work at decent/top rates. Reading up on it/courses/home tinkering isn't enough to get people in to mainstream IT contracts let alone emerging tech.

A few people might get lucky having a hand in somewhere but I don't think there is going to be much work out there for your average contractor bod that fancies a change of tech for awhile yet. That's not to say start doing everything you can to prepare but its going to be difficult to gain the skills and experience you need to sell back to a client in the short term.

You make a valid point about the contract element and the challenges that AI contractors may face, especially in the short term. AI is indeed an emerging technology with a rapidly evolving landscape, and building internal capabilities is a common strategy for companies looking to integrate AI into their long-term plans. This could potentially reduce the immediate demand for external AI contractors, as companies may prioritize upskilling their existing workforce or hiring permanent employees to ensure sustained expertise.

Here are some factors to consider when it comes to AI contracting work in the short term:

1. Specialized Skills and Experience: As you mentioned, gaining expertise in AI typically requires more than just reading up on the subject or taking courses. It often demands hands-on experience, exposure to real-world AI projects, and a deep understanding of AI concepts. For contractors looking to enter the AI market, acquiring these specialized skills may take time and dedication.
2. Evolving Job Market: The AI job market can be dynamic, and the demand for specific AI skills may fluctuate based on industry trends, technological advancements, and the maturity of AI use cases. Contractors must be adaptable and keep up with the latest developments in the AI field to stay relevant and attractive to potential clients.
3. Building a Portfolio: Demonstrating relevant AI experience through a portfolio of successful projects is crucial for AI contractors. Gaining the necessary experience to build a strong portfolio might take time, and contractors may need to consider contributing to open-source projects, participating in hackathons, or collaborating with AI-focused communities to gain visibility and credibility.
4. Niche Expertise: While some entry-level AI contracting work may be available, more advanced and complex AI projects may require niche expertise. Contractors with domain-specific knowledge or experience in particular industries might find unique opportunities that others might not.
5. Networking and Branding: Building a reputation and network within the AI community can significantly impact a contractor's chances of securing high-value AI contracts. Engaging in conferences, meetups, and online forums can help establish connections and showcase expertise.
6. Value Proposition: To attract clients, contractors need to offer a clear value proposition that sets them apart from the competition. Understanding the specific challenges faced by potential clients and demonstrating how AI expertise can address those challenges is crucial.

In summary, you are right that AI contracting work may be challenging to secure in the short term for your average contractor who is relatively new to AI. It will take time and effort to gain the necessary skills, experience, and network to become a sought-after AI contractor. However, for those willing to invest in their AI expertise and continuously adapt to the evolving AI landscape, there will likely be increasing opportunities as AI becomes more pervasive in various industries and applications.

[Lance]

Several years back, the Next Big Thing in financial services (and elsewhere, probably) was blockchain. Everyone thought it was going to revolutionise the industry. It certainly had the potential to significantly improve it in many ways, but it's really taking its time to get there.

I see AI in similar terms. It'll be great for some applications and will almost certainly be adopted more widely than blockchain, but unless a company can see a major payoff in terms of reduced costs, increased revenue or commercial advantage, they're unlikely to do much more than play with it.

As with most things, I'll probably be proven wrong though

Your perspective on AI and its potential impact on the financial services industry is astute. AI, like blockchain, has garnered significant attention and excitement for its transformative potential across various sectors. However, the actual pace of adoption and implementation of AI solutions may not always align with the initial hype.

The comparison to blockchain is relevant because both AI and blockchain are technologies that have the potential to revolutionize industries, but their adoption trajectories have been different:

1. Complexity and Maturity: Both AI and blockchain technologies can be complex and require careful consideration and expertise for successful implementation. However, AI has seen more widespread adoption in certain areas due to its maturity, availability of pre-built models and tools, and practical applications in diverse use cases.
2. Use Cases and Business Value: Just as you mentioned, the adoption of AI (or any technology) is often driven by the tangible value it brings to businesses. Companies are more likely to invest in AI when they see a clear payoff in terms of reduced costs, increased efficiency, better customer experience, or competitive advantage. AI is well-suited for tasks involving pattern recognition, prediction, optimization, and automation, making it appealing for various financial services applications.
3. Data Availability and Integration: AI's effectiveness often depends on the availability and quality of data. In the financial services sector, access to large and diverse datasets is often crucial for developing accurate and reliable AI models. Integrating AI with existing systems and processes can also be challenging, making it a more gradual process for some organizations.
4. Regulatory and Security Concerns: Financial services are heavily regulated, and any new technology implementation must comply with industry regulations and data security standards. This can slow down the adoption of AI solutions, as companies need to ensure that they meet all necessary requirements.
5. Cultural Shift: Adopting new technologies like AI can also require a cultural shift within organizations. There might be resistance to change, and stakeholders may need time to embrace and understand the benefits of AI fully.

Despite these challenges, AI is making significant strides in financial services and other industries. As AI technologies continue to evolve and mature, and as companies witness successful use cases and best practices from early adopters, we can expect broader adoption of AI over time. The successful integration of AI into some sectors will likely pave the way for more applications and innovations, but it may take longer for its full potential to be realized in certain contexts.

As with any transformative technology, there can be both overestimation and underestimation of its impact. It's essential to approach AI with a realistic outlook, acknowledging its potential while understanding the complexities and challenges involved in its adoption. Time will indeed reveal the true extent of AI's impact and how it reshapes various industries, including financial services.

[Lance]

CAVEAT::: I know f*** all about this subject. I just pasted the question and responses into chat-gpt 3.5

I guess it's passed the Turing test