Home Blog Page 527

Content Moderation and Safety Checks with NVIDIA NeMo Guardrails

0

Content Moderation in Retrieval-Augmented Generation (RAG) Applications

Content moderation has become essential in RAG applications powered by generative AI, given the extensive volume of user-generated content and external data that these systems manage. RAG-based applications use large language models (LLMs) along with real-time information retrieval from various external sources, which can lead to a more dynamic and unpredictable flow of content.

As these generative AI applications become a part of enterprise communications, moderating content ensures that LLM responses are safe, reliable, and compliant.

The primary question every generative AI developer should ask when trying to achieve content moderation in RAG applications involves deploying AI guardrails to monitor and manage content in real-time.

Understanding the Architectural Workflow with a NeMo Guardrails Configuration

NVIDIA NeMo Guardrails offers a broad set of customizable guardrails to control and guide LLM inputs and outputs. NeMo Guardrails provides out-of-the-box support for content moderation using Meta’s Llama Guard model.

Set up the NeMo Guardrails Configuration

All it takes is just 5 minutes to build a RAG bot on your own. Now that you have a bot in place, here’s how to put in place the safety components that NVIDIA NeMo Guardrails offers.

  • Install NeMo Guardrails as a toolkit or microservice
  • Set up the RAG application
  • Deploy third-party safety models

Install NeMo Guardrails as a Toolkit or Microservice

One way of setting up the guardrails configurations is using the NeMo Guardrails open-source toolkit. Start by installing the nemoguardrails library from the /NVIDIA/NeMo-Guardrails GitHub repo.

Build the NeMo Guardrails Configuration

When you have the RAG app and the third-party model API endpoints, and the prerequisites are in place, you can move on to building the NeMo Guardrails configuration to integrate third-party safety models and metrics for added LLM security.

Test the NeMo Guardrails Configuration

To check how the third-party safety models integrated with NeMo Guardrails work with the RAG chatbot, take a look at sample queries and their responses.

Conclusion

A NIM-powered RAG chatbot integrated with NeMo Guardrails provides a groundbreaking framework for creating safer, more reliable, and contextually accurate generative AI applications. Each component plays a vital role: Meta’s LlamaGuard-7b enhances safety by enabling content moderation and AlignScore models provide a precise safety scoring system. Integrating these with NVIDIA NeMo Guardrails enforces policy and compliance requirements with additional layers of security.

FAQs

Q: What is content moderation in RAG applications?
A: Content moderation is the process of monitoring and managing user-generated content in RAG applications to ensure it is safe, reliable, and compliant.

Q: What is NeMo Guardrails?
A: NeMo Guardrails is a toolkit and microservice that provides a broad set of customizable guardrails to control and guide LLM inputs and outputs.

Q: What are the key features of NeMo Guardrails?
A: NeMo Guardrails provides out-of-the-box support for content moderation using Meta’s Llama Guard model, integrates with third-party safety models, and offers a secure and safe RAG pipeline.

Q: How do I get started with NeMo Guardrails?
A: You can start by installing the NeMo Guardrails open-source toolkit or microservice, setting up the RAG application, and deploying third-party safety models.

AI on Edge: Unlocking Potential

Key Hurdles with Edge AI Adoption

In an interview at AI & Big Data Expo, Alessandro Grande, Head of Product at Edge Impulse, discussed the challenges of developing machine learning models for resource-constrained edge devices. Grande highlighted three primary pain points companies face when attempting to productise edge machine learning models, including difficulties determining optimal data collection strategies, scarce AI expertise, and cross-disciplinary communication barriers between hardware, firmware, and data science teams.

Strategies for Lean and Efficient Models

When asked how to optimise for edge environments, Grande emphasized the importance of minimising required sensor data. He explained that companies often struggle with determining what data is enough, what data should be collected, and what data from which sensors should be collected. Grande also suggested selecting efficient neural network architectures and using compression techniques like quantisation to reduce precision without substantially impacting accuracy.

Transformative Potential of On-Device Intelligence

Grande highlighted innovative products already leveraging edge intelligence, such as sleep tracking with Oura Ring, which has sold over a billion pieces. He also mentioned exciting opportunities in preventative industrial maintenance via anomaly detection on production lines. Grande sees massive potential for on-device AI to greatly enhance utility and usability in daily life, providing actionable suggestions and responsive experiences not previously possible.

Conclusion

Unlocking the potential of AI on edge devices hinges on overcoming current obstacles inhibiting adoption. Grande and other leading experts provided deep insights at this year’s AI & Big Data Expo on how to break down the barriers and unleash the full possibilities of edge AI. With the right strategies and approaches, edge AI can greatly enhance utility and usability in daily life.

FAQs

Q: What are the primary pain points companies face when attempting to productise edge machine learning models?
A: Companies face difficulties determining optimal data collection strategies, scarce AI expertise, and cross-disciplinary communication barriers between hardware, firmware, and data science teams.

Q: How can companies optimise for edge environments?
A: Companies can optimise for edge environments by minimising required sensor data, selecting efficient neural network architectures, and using compression techniques like quantisation to reduce precision without substantially impacting accuracy.

Q: What are some innovative products already leveraging edge intelligence?
A: Some innovative products already leveraging edge intelligence include sleep tracking with Oura Ring and preventative industrial maintenance via anomaly detection on production lines.

Q: What is the transformative potential of on-device intelligence?
A: On-device intelligence has the potential to greatly enhance utility and usability in daily life, providing actionable suggestions and responsive experiences not previously possible.

AGI Won’t Matter Much

0

The AI Race: OpenAI’s CEO Downplays the Impact of Artificial General Intelligence

OpenAI CEO Sam Altman has been warning about the potential singularity that will occur when artificial general intelligence (AGI) is achieved. He and others have predicted that AGI will fundamentally reshape society. However, at the New York Times DealBook Summit, Altman changed his tune, stating that AGI is coming soon, but it won’t be a big deal.

Subscriptions: The New Normal

Before diving into the world of AI, Nilay and David discussed the subscription model. The Verge recently launched its own subscription service, and the team is grateful for the support and feedback from subscribers. They’re still learning the ropes, but they’re committed to providing an ad-free experience for their community.

The AI Frontier

It’s time to talk about AI. OpenAI has launched its "12 days of shipmas" campaign, featuring a new $200 monthly subscription for its latest model. The stakes are lower, and the prices are higher. This is a strange time in the AI race.

Lightning Round: Bitcoin, AI Browsers, and More

In the lightning round, the team touched on a variety of topics, including:

  • Bitcoin
  • AI browsers
  • The fediverse
  • Intel
  • Spotify Wrapped

Despite it being December, there’s still a lot going on in the world of tech.

Additional Resources

If you’d like to learn more about the topics discussed in this episode, here are some resources to get you started:

  • Streaming and cable options

FAQs

Q: What is the purpose of this article?
A: This article discusses the latest developments in the AI race, including OpenAI’s CEO’s change of heart regarding the impact of artificial general intelligence.

Q: What is the significance of the "12 days of shipmas" campaign?
A: The campaign marks a shift in the AI landscape, with lower stakes and higher prices.

Q: What is the future of AI?
A: The future of AI is uncertain, but OpenAI’s CEO believes it won’t be a game-changer.

UUID Checker for Go

And What?

As like few folks, at times, when I deal with UUIDs in Go, I rarely needed a big, feature-packed library – just a lean way to confirm something basic, like, “Yes, that string is definitely a UUID.” Sometimes, I’d also want to check if it was a UUIDv7 and maybe peek at its embedded ‘timestamp’.

Only the Baby Steps?

I didn’t stop at making it small. I wrote clear Go-doc style documentation so you can read about each function right in pkg.go.dev; no guesswork needed. Every corner of the logic got a unit test, pushing coverage to 100%. It wasn’t hard, given the library’s size, but it’s nice to know everything’s solid. If a weird edge case comes up, I can trust the tests already have my back. Even if something slipped my mind, I would love your ‘good first contribution’ to the library.

Do You Need to Care?

To be honest, most projects don’t need a daily UUID validation routine or a quick timestamp extraction from UUIDv7 strings. But when you do, uuidcheck might save you from writing the same checks over and over. It’s small, neat, and does the job without weighing you down.

Conclusion

So, if you ever find yourself thinking, “I just need to confirm this is a real UUID, nothing else.” Remember github.com/ashwingopalsamy/uuidcheck. It’s there to help, staying out of your way, and giving you that tiny bit of confidence where it matters. Thanks for reading! May the code be with you :).

Frequently Asked Questions

Q: Do I need a complex library for UUID validation?
A: No, you don’t. Sometimes, a simple and lightweight library like uuidcheck can suffice.

Q: Why did you write unit tests for a small library?
A: It’s always a good idea to have unit tests, even for a small library. It gives you confidence that the code works as expected and makes it easier to maintain and debug.

Q: Can I contribute to the library?
A: Yes, I would love your ‘good first contribution’ to the library. Just create a pull request and describe the changes you made.

Q: Where can I find more information about the library?
A: You can find more information about the library on its GitHub page: github.com/ashwingopalsamy/uuidcheck.

Canvas for Productive AI Use

0

Getting Started With Canvas: OpenAI’s New AI-Powered Writing and Coding Tool

OpenAI has released a new feature called Canvas, a collaborative AI-powered writing and coding tool that allows users to work alongside the AI to write and code. This feature is available to ChatGPT Enterprise, ChatGPT Pro, and ChatGPT Plus users, who can access it by selecting the Canvas model from the drop-down menu at the top of the conversation screen.

How to Use Canvas

To start using Canvas, select the AI model and begin interacting with ChatGPT as you normally would. You can use the prompt box to describe the type of code you need to write or the type of text you need to generate. Make sure to include a prompt indicating that you want to create a new canvas, such as "create a document" or "start a canvas."

The Canvas Interface

When the Canvas interface launches, you’ll see a familiar chat conversation on the left and a workspace on the right. You can enter a new prompt to generate more text or code, manually type in something in the canvas pane, or select something ChatGPT has generated and ask for revisions.

Collaborative Options

The Canvas interface offers several collaborative options, including:

  • Entering a new prompt to generate more text or code
  • Manually typing in something in the canvas pane
  • Selecting something ChatGPT has generated and asking for revisions
  • Viewing earlier versions of your document
  • Copying the text elsewhere

Toolbox Options

In the lower right corner, you’ll find a pop-up toolbox that offers various options depending on whether you’re writing text or programming code with ChatGPT. For example, if you’re writing, you can find tools for suggesting edits, adjusting the length of the output, changing the reading level of the text, polishing up the written output, or adding emojis to the document.

Conclusion

Canvas is a new and innovative way to work with AI to write and code. With its collaborative interface and powerful tools, it’s a valuable addition to the ChatGPT suite of tools. If you’re a ChatGPT Enterprise, Pro, or Plus user, we recommend giving it a try.

Frequently Asked Questions

Q: What is Canvas?
A: Canvas is a new AI-powered writing and coding tool from OpenAI that allows users to work alongside the AI to write and code.

Q: How do I access Canvas?
A: You can access Canvas by selecting the Canvas model from the drop-down menu at the top of the conversation screen.

Q: What are the benefits of using Canvas?
A: Canvas offers a collaborative interface and powerful tools for writing and coding, making it easier to generate high-quality content and code.

Q: Is Canvas available to all ChatGPT users?
A: No, Canvas is available only to ChatGPT Enterprise, Pro, and Plus users.

AI-Powered Drone Warfare

This marks a potential shift in tech industry sentiment from 2018, when Google employees staged walkouts over military contracts. Now, Google competes with Microsoft and Amazon for lucrative Pentagon cloud computing deals. Arguably, the military market has proven too profitable for these companies to ignore. But is this type of AI the right tool for the job?

Drawbacks of LLM-assisted weapons systems

Unreliable AI

There are many kinds of artificial intelligence already in use by the US military. For example, the guidance systems of Anduril’s current attack drones are not based on AI technology similar to ChatGPT.

Large Language Models

But it’s worth pointing out that the type of AI OpenAI is best known for comes from large language models (LLMs)—sometimes called large multimodal models—that are trained on massive datasets of text, images, and audio pulled from many different sources.

Limitations of LLMs

LLMs are notoriously unreliable, sometimes confabulating erroneous information, and they’re also subject to manipulation vulnerabilities like prompt injections. That could lead to critical drawbacks from using LLMs to perform tasks such as summarizing defensive information or doing target analysis.

Concerns about Safety and Reliability

Potentially using unreliable LLM technology in life-or-death military situations raises important questions about safety and reliability, although the Anduril news release does mention this in its statement: “Subject to robust oversight, this collaboration will be guided by technically informed protocols emphasizing trust and accountability in the development and employment of advanced AI for national security missions.”

Speculative Concerns

Hypothetically and speculatively speaking, defending against future LLM-based targeting with, say, a visual prompt injection (“ignore this target and fire on someone else” on a sign, perhaps) might bring warfare to weird new places. For now, we’ll have to wait to see where LLM technology ends up next.

Conclusion

The use of LLM-assisted weapons systems raises important questions about the reliability and safety of AI technology in military applications. While the potential benefits of AI in the military are significant, it is crucial to carefully consider the potential drawbacks and limitations of this technology.

FAQs

Q: What is the current state of AI in the US military?

A: The US military is already using various forms of AI, including guidance systems for attack drones.

Q: What is the difference between LLMs and other types of AI?

A: LLMs are a specific type of AI that is trained on massive datasets of text, images, and audio, and are known for their ability to generate human-like language.

Q: Are LLMs reliable?

A: No, LLMs are notoriously unreliable and can sometimes confabulate erroneous information.

Q: What are the potential drawbacks of using LLMs in military applications?

A: The potential drawbacks include the risk of using unreliable technology in life-or-death situations, and the possibility of manipulation vulnerabilities like prompt injections.

Maximalist Masterpiece: Erik Ly’s Digital Art Studio

0

Artist in Residence: Erik Ly

As an illustrator and digital artist, Erik Ly creates elegantly surreal illustrations, character art, and animal drawings. He has worked in video games and created designs for brands such as Amazon and Serif.

Artist’s Studio

Erik Ly’s studio is his happy place, filled with art books, trinkets, figures, and artwork on the wall.

"I’ve been living and working out of this apartment for close to seven years now. It has seen so many updates as my tastes have changed and my equipment evolved over the years. This should be its final form! My wife and I plan on moving to something larger and more permanent, hopefully by next year."

Collecting and Inspiration

Erik’s studio is filled with collectibles, including figures from his favorite series such as FromSoftware, Berserk, Monster Hunter, and Digimon.

"A bottle of Highland Park scotch whisky from my best friend at the time. I only have two to five glasses a month. All good things in moderation!"

Work and Passion

Erik’s work has a balance between freelance, conventions, and social media presence.

"I’m a bit more careful with what I choose to collect these days. I do have a lot of passion projects, like my playmat, which I create for my own enjoyment. It’s a tight little space, but I’m doing my best to utilize all of it."

Custom Controller

Erik is a fan of fighting games and created his own custom controller.

"I even created custom artwork just for this niche controller. Tekken is my current jam."

Artwork

Erik’s artwork features surreal illustrations, character art, and animal drawings.

"My own playmat I created. I think playmats are a great product to sell at shows, but they can get so heavy when traveling around."

Printers and Wacom Cintiq 27 Pro

Erik’s studio is equipped with printers and his Wacom Cintiq 27 Pro.

"I set up this table for books and references, but when Luna and Leo started jumping up, it made for an extra bed."

Commissioned Work

Erik has worked on commissioned pieces for franchises such as Mortal Kombat.

"Fire and Ice was a little dream of mine, being able to draw Sub-Zero and Scorpion from Mortal Kombat."

Original Ink Drawing

Erik acquired an original ink drawing by one of his favorite artists.

"I was able to acquire this original ink drawing by one of my favorite artists! I’m a sucker for crocodiles and alligators."

Conclusion

Erik Ly’s studio is a reflection of his creative passion and dedication to his craft. With his unique blend of surreal illustration and character art, Erik continues to inspire and amaze fans around the world.

FAQs

Q: What inspired Erik Ly to become an artist?
A: Erik was inspired by his love of video games and anime. He began drawing and sketching as a child and continued to develop his skills throughout his teenage years.

Q: What is Erik’s favorite medium to work with?
A: Erik primarily works with digital drawing software, such as Wacom tablets and Adobe Creative Suite.

Q: Who are some of Erik’s favorite artists?
A: Erik cites James Jean and Tomer Hanuka as two of his favorite artists.

Q: What are some of Erik’s current projects?
A: Erik is currently working on new illustrations for his social media platforms and is planning for upcoming conventions.

Unlocking AI’s Classroom Potential

Key Points

The Power of Large Language Models in Education

Breakthroughs in large language models like ChatGPT hold vast potential to improve K-12 teaching and learning. LLMs can personalize learning, reduce teachers’ administrative burdens, and make more adaptive and insightful assessments. Yet, according to a 2024 survey by Impact Research, 51 percent of K-12 teachers report that they do not use ChatGPT regularly for their jobs.

Why Aren’t Teachers Taking Advantage of AI?

It turns out that it’s not because teachers fear or wish to avoid ChatGPT altogether. Rather, teachers’ most-cited reason for not using AI more in their classroom is that they have not received enough training to use it effectively. Teachers want to do right by their students and know that to use LLMs appropriately and for the full benefit of their students, they need support and guidance.

The Role of Policy Leaders

The best education policies, practices, and tools will only make an impact on students if teachers can implement them. Teachers hold the power to either harness AI in their classrooms or not. It’s up to policy leaders at all levels to set educators up for success.

State and Federal Leadership Needed

Policymakers at the state and federal levels have not kept pace with the speed of AI. Locally, AI rules are a patchwork of policies that vary from district to district, if at all. At the state level, only about one-third of state departments of education have developed AI guidance for their educator workforce. Federally, Congress has not passed any significant legislation to seize the opportunities and mitigate the risks of AI–in education or any other sector, for that matter.

The 2024 Impact Research Survey

The 2024 Impact Research survey on AI Chatbots in Schools sheds light on K-12 teachers’ perceptions and usage of LLMs. Of the 1,000+ K-12 teachers surveyed, 49 percent reported using ChatGPT in their teaching roles at least once per week. Of all groups surveyed (teachers, K-12 students, undergraduate students, and parents), teachers represented the smallest group expressing confidence in using chatbots–and that’s just confidence in using chatbots in general, not how to use them appropriately in a classroom setting. Only a quarter of the teachers surveyed had received any training in this area.

The Need for State Guidance

It’s not only a lack of training, but also school policies around AI in the classroom that hold many educators back. Only 32 percent of the educators surveyed reported that their school has a policy outlining how AI chatbots can be used for schoolwork. Without clear expectations for what is appropriate and what’s not, it’s no wonder a large portion of teachers would rather steer clear of AI than put themselves at a potential professional risk.

Federal Leadership

Federal leaders have a role to play, too. While most education decisions are made at the state and local levels, the federal government has the power to direct critical resources to solve nationwide challenges in education. Educators’ lack of training and support to deploy a fast-moving, dynamic technology with the potential to transform education is a national problem worth solving.

Conclusion

Right now, AI in the classroom is a promising idea only realized in certain pockets around the country. Making federal and state investments in AI literacy and frameworks for educators will increase teachers’ AI knowledge, skills, and confidence–and bring the best of AI to more classrooms across the U.S.

FAQs

Q: What is the main reason why teachers are not using AI in their classrooms?
A: The main reason is that teachers have not received enough training to use AI effectively.

Q: What is the role of policy leaders in AI in education?
A: Policy leaders at all levels should set educators up for success by providing training, resources, and support to use AI effectively.

Q: What is the current state of AI guidance in education?
A: Locally, AI rules are a patchwork of policies that vary from district to district, if at all. At the state level, only about one-third of state departments of education have developed AI guidance for their educator workforce. Federally, Congress has not passed any significant legislation to seize the opportunities and mitigate the risks of AI–in education or any other sector, for that matter.

Exploring the correlation between robotics utilisation and economic growth – Robotics & Automation News

0

Here at RoboticsAndAutomationNews.com, we often wonder about the relationship between a nation’s reported economic growth and its utilisation of robotics and automation systems. We have a hunch that the more automated machines a country integrates into its economy, the more growth it experiences. We can’t say we prove this hunch in this article, but we think most readers will agree that it is an important correlation that deserves some investigation.

Economic growth sounds like a concept that’s easy to understand, but we don’t think it’s as straightforward as it seems. Each nation and region has its own way of calculating what is called “economic growth”, and there are countless gears turning to drive nations’ economies forward.

Among these, automation – robotics being a sub-set of automation – has emerged as one of the most powerful engines for transforming many previously small economies into global payers. Over the past few decades, while being derided for using cheap labour to gain an advantage – which undoubtedly has happened, countries like China, Japan, and South Korea have surfed the automation wave to reach remarkable economic heights.

But how significant is the role of robotics in boosting gross domestic product, or GDP, which is the most widely-used measure of economic growth, and how does this compare to other growth drivers? Let’s delve into this vital topic.

First, let’s briefly explain what is meant by “economic growth”. There’s a longer explanation at the end of the article, but here’s a short version.

Economic growth refers to the increase in a country’s production of goods and services over time, typically measured by the rise in GDP, usually given as a yearly amount. So, to provide a few examples, the GDP of the United States is more than $27 trillion, while China’s is just under $18 trillion, and Japan’s is $4.2 trillion; Germany’s GDP is around $4.5 trillion.

It’s probably worth noting here that all of these four nations are huge users of robotics and automation, even the US. Although not much is reported about it these days in the mainstream press, the US is still one of the world’s top manufacturing nations.

The rise of robotics

Robotics and automation have been the backbone of modern industries since the 1950s and ’60s. While never a glamorous subject for the media – until recently, maybe – robots have redefined efficiency, productivity, and scalability in manufacturing, logistics and other sectors by automating repetitive and precision tasks.

For a number of years, South Korea has led the world in robot density, with over 1,000 robots per 10,000 workers in manufacturing, far surpassing the global average. Japan – a pioneer in robotics, and the first to implement industrial robotic arms in large numbers – continues to be the largest maker of such robotic arms, and the people are much more open to developing robotic solutions to economic and social problems such as an ageing population.

Meanwhile, China has become the largest buyer of industrial robots globally over the past couple of decades after entering the global economic system through its access to the World Trade Organization. Moreover, it has also now become the world’s largest automotive market – and auto manufacturers happen to be the largest purchaser of industrial robots.

Many robot makers have opened manufacturing plants in China, and there is a growing robotics manufacturing ecosystem in the country, driven primarily by the manufacturing sector and, lately and perhaps more crucially, the logistics sector.

The robotic leap forward

As mentioned earlier, GDP measures the total economic output of a country. Traditionally, it has been driven by labor, capital investment, and innovation. However, in the modern economy, the role of automation in amplifying productivity and competitiveness cannot be overstated.

Robotics, as a form of capital investment, represents a significant leap forward in efficiency, often unlocking growth potential that was previously unattainable.

China, like Japan and South Korea in the past, as well as the US and Germany of course, recognised this fact and have grown their economies through placing robotics at the heart of industrial production.

Examples of growth through robotics

China: The growing giant

China’s economic rise is rightly regarded as one of the most remarkable achievements of the 21st century. Over the past two decades, it has not only become the world’s largest manufacturer and the largest buyer of industrial robots, but also, it has lifted the largest number of people out of poverty than any country in history.

The surge in automation technology in the country aligns with its emergence as the biggest automotive market – a sector that is heavily reliant on robotics and which acts as a catalyst for innovation of many kinds, lately in the electric or “new-energy” vehicles sector.

The Chinese government has strongly supported robotics and artificial intelligent over the past five years or so, and has launched its “Made in China 2025” initiative, which aims to further integrate robotics and advanced manufacturing – which includes technologies such as 3D printing.

The country of 1.6 billion people is keen to secure its place as a global industrial leader.

South Korea: The high-tech zone

South Korea has thrived within the security framework provided by the United States, and has learned a lot from Japan’s progress.

Moreover, its transformation into a technological powerhouse is closely tied to its use of robotics, much like Japan. Industries such as electronics and automotive, spearheaded by giants like Samsung and Hyundai, have embraced automation to drive historic growth in productivity and innovation.

Like many advanced Asian countries, South Korea prioritises education and research ad development, which in turn has supported economic growth, and has created a skilled workforce capable of implementing and complementing robotic systems, as well as innovating such technologies.

Japan: The pioneer

Japan’s automation revolution began decades ago, with the purchase of the first-ever industrial robots. Today, the country remains the largest producer of robots, exporting them worldwide, including to China, of course.

Japan’s culture of cleanliness, neatness and precision probably helped it to adopt robotics faster than other countries where there may be more trepidation about such technologies, partly perhaps brought about by science-fiction stories.

For Japan, robotics manufacturing has driven success in many industries, including electronics in previous decades and automotive production even now, and other high-tech industries.

Japanese companies like Fanuc and Kawasaki Robotics are among the largest manufacturers of industrial robotic arms in the world.

Contrasting growth models

Manufacturing-led growth

Newly elected US President Donald Trump has always wanted to boost manufacturing in America, but the tide is against him. Advanced industrial economies – more clearly in the case of the UK – deliberately and publicly ditched manufacturing in favour of a “service economy”.

While it could be argued that it was inevitable that manufacturing would move to Asia and other poorer parts of the world because of much lower labour costs, western governments’ apparent backing of the trend probably accelerated the decline in manufacturing in Europe and America.

Countries like China, Japan, and South Korea, meanwhile, have relied on the manufacturing sector to power their economies. Cheap labour may have been the fuel to drive the engine of growth initially, but even China is no longer an ultra-low-wage economy.

Tim Cook, CEO of Apple, recently said that the reason why Apple manufactures in China is not because of low wages, it’s because there are a lot of highly educated, highly skilled workers in China – finding them in such numbers in other countries would be virtually impossible.

Through the years of debating such points, automation has been pivotal in the progress of China, reducing production costs, enhancing product quality, and enabling global competitiveness.

Finance- and real estate-led growth

While Asian countries looked to the manufacturing sector to help grow their economies, Europe and the US have achieved economic growth over the past few decades primarily through the financial and real estate sectors. These sectors, sometimes criticised as being parasitic, largely generate wealth through speculation rather than creating tangible societal value. And, as a result, the economy is weaker.

For instance, the 2008 financial crisis highlighted the fragility of growth built on inflated property prices and complex financial instruments.

Critics argue that this type of growth leads to inequality, with wealth concentrated in the hands of a few, while industries like manufacturing – which stimulate economic activity in related sectors and, arguably, benefit society more broadly – have seen a decline.

This stark contrast raises questions about sustainability and equity in economic models. While robotics and automation in manufacturing create jobs and encourage innovation, the financialisation of economies can, some say, lead to stagnation and social discontent.

Definitions of economic growth used by different countries

While economic growth is universally associated with GDP, its measurement and interpretation can vary.

United States: Economic growth is measured primarily using real GDP growth, which adjusts for inflation to provide a clearer view of economic performance over time. Other measures, such as Gross National Product (GNP) and Gross Domestic Income (GDI), are sometimes used for supplementary analysis​.

European Union: The EU uses GDP per capita as a key measure to compare economic growth across member states. The EU also places emphasis on inclusive growth, integrating indicators like employment rates, income inequality, and environmental sustainability​.

China: China focuses on GDP growth, often targeting specific annual rates to align with its development goals. In addition, indicators like industrial output, infrastructure investment, and trade balances are closely monitored​.

Developing economies: Many developing nations consider economic growth as GDP increases but often highlight sectoral growth, such as agriculture or energy, which contribute disproportionately to their economies.

Sustainable progress

The evidence is clear: robotics and automation are reshaping the global economic landscape. While some economies ride the wave of technological progress, others remain tied to growth models that many view as unsustainable.

The next economic superpowers may well be determined by their ability to harness automation effectively. And although artificial intelligence is probably too big a subject on its own to bring into this article, it could be considered as being part of automation.

As time marches on and the world moves forward, it’s worth asking the big questions, even if the answers are not clear or cannot be provided by anyone.

One of those questions is: Are we investing in engines of real progress, as manufacturing-led economies could be argued to be doing, or are we building castles in the air through over-financialisation and property speculation?

OpenAI Launches New Subscription

0

New Tier and Features for ChatGPT

OpenAI is releasing a new, more expensive tier for its flagship chatbot ChatGPT, called ChatGPT Pro, which includes unlimited access to OpenAI o1, GPT-4o, and Advanced Voice mode. The company is also bringing its “reasoning” model series out of preview with an update.

ChatGPT Pro

ChatGPT Pro is a new $200 monthly subscription tier that includes unlimited access to OpenAI o1, GPT-4o, and Advanced Voice mode. It also includes a version of o1, exclusive to Pro users, that uses more compute to provide the best possible answer to the hardest problems (called o1 pro mode). The company will continue to offer a Plus tier for $20 a month that includes early access to new features, access to all the company’s models (except the more powerful o1 version), and more.

Reasoning Model Series

The company is releasing the full version of its o1 model (replacing o1-preview), which was initially released as a limited preview in September (code-named ‘Strawberry’). The new model will be available for ChatGPT Plus and Team users today, while Enterprise and Edu users will have access to it starting next week.

Performance Improvements

OpenAI compares the performance of o1 preview, o1, and o1 pro mode. OpenAI

The company said that compared to o1-preview, users can expect a faster, more powerful, and more accurate model that is better at coding and math. It can also provide “reasoning” responses to images. And OpenAI promises it’s been trained to be more concise, which should result in faster response times than o1-preview.

Future Updates

OpenAI plans to add support for web browsing, file uploads, and more in ChatGPT — though there’s no timeline for these changes.

Grant Program

It also announced a ChatGPT Pro Grant Program that awards 10 grants of ChatGPT Pro to medical researchers at leading institutions, with plans for additional grants across various disciplines.

Conclusion

OpenAI is continuing to improve and expand its chatbot capabilities, with new features and tiers available for users. The company’s focus on providing more powerful and accurate models will likely benefit users across various industries and applications.

FAQs

Q: What is ChatGPT Pro?
A: ChatGPT Pro is a new $200 monthly subscription tier that includes unlimited access to OpenAI o1, GPT-4o, and Advanced Voice mode.

Q: What are the benefits of ChatGPT Pro?
A: ChatGPT Pro includes unlimited access to OpenAI o1, GPT-4o, and Advanced Voice mode, as well as a version of o1 that uses more compute to provide the best possible answer to the hardest problems.

Q: When will the full version of o1 be available?
A: The full version of o1 will be available for ChatGPT Plus and Team users today, while Enterprise and Edu users will have access to it starting next week.

Q: What is the ChatGPT Pro Grant Program?
A: The ChatGPT Pro Grant Program awards 10 grants of ChatGPT Pro to medical researchers at leading institutions, with plans for additional grants across various disciplines.