Why smart AI regulation is vital for innovation and U.S. leadership Gary Shapiro 9 hours Gary Shapiro Contributor Share on Twitter Gary Shapiro is the president and CEO of the Consumer Technology Association and a New York Times bestselling author of the book Ninja Future: Secrets to Success in the New World of Innovation. More posts by this contributor
New approach to state drone laws balances privacy and innovation
The biggest threat to drone innovation is a group you’ve never heard of
As a teenager, I immersed myself in science fiction. While the visions of many films and novels haven’t come to pass, I’m still amazed by legendary writer Isaac Asimov’s ability to imagine a future of artificial intelligence and robotics. Now, amid all the hype around generative AI and other AI tools, it’s time for us to follow Asimov’s lead and write a new set of rules.
Of course, AI rules for the 21st century won’t be quite as simple as Asimov’s three rules of robotics (popularized in I, Robot). But amid anxiety around the rise of AI tools and a misguided push for a moratorium on advanced AI research, industry can and should be pushing for rules for responsible AI development. Certainly, the past century’s advances in technology have given us plenty of experience in evaluating both the benefits of technological progress and the potential pitfalls.
Technology itself is neutral. It’s how we use it – and the guardrails we set up around it – that dictate its impact. As humans, harnessing the power of fire allowed us to stay warm and extend food storage time. But fire can still be destructive.
Think of how the recent Canadian wildfires threatened lives and property in Canada and damaged U.S. air quality. Nuclear power in the form of atomic bombs killed thousands in Japan during WWII, but nuclear energy lights up much of France and powers U.S. aircraft carriers.
We’re at a pivotal moment for the future of an amazing, complex and consequential technology. We can’t afford to let other countries take the lead.
In the case of AI, new tools and platforms can solve big global problems and create valuable knowledge. At a recent meeting of Detroit-area Chief Information Officers, attendees shared how generative AI is already speeding up time-to-market and making their companies more competitive.
Generative AI will help us “listen” to different animal species. AI will improve our health by supporting drug discovery and disease diagnosis. Similar tools are providing everything from personalized care for elders to better security for our homes. More, AI will improve our productivity, with a new study by McKinsey showing generative AI could boost the global economy by $4.4 trillion annually.
With all this possibility, can such an amazing technology also be bad? Some of the concerns around AI platforms are legitimate. We should be concerned about the risk of deep fakes, political manipulation, and fraud aimed at vulnerable populations, but we can also use AI to recognize, intercept and block harmful cyber intrusions. Both barriers and solutions may be difficult and complex, and we need to work on them.
Some may also be simple; we already see schools experimenting with oral exams to test a student’s knowledge. Addressing those issues head-on, rather than sticking our heads in the sand with a pause on research that would be impossible to enforce and ripe for exploitation by bad actors, will position the United States as a leader on the world stage.
While the U.S. approach to AI has been mixed, other countries seem locked in to a hyper-regulatory stampede. The EU is on the precipice of passing a sweeping AI Act that would require companies to ask permission to innovate. In practice, that would mean that only the government or huge companies with the finances and capacity to afford the certification labyrinth covering privacy, IP, and a host of social protection requirements could develop new AI tools.
A recent study from Stanford University also found that the EU’s AI Bill would bar all of the currently existing large language models, including OpenAI’s GPT-4 and Google’s Bard. Canadian lawmakers are moving forward an overly broad AI bill that could similarly stifle innovation. Most concerning, China is rapidly pursuing civil and military AI dominance through massive government support. More, it has a different view of human rights and privacy protection which may help its AI efforts but is antithetical to our values. The U.S. must act to protect citizens and advance AI innovation or we will be left behind.
What would that look like? To start, the U.S. needs a preemptive federal privacy bill. Today’s patchwork of state-by-state rules mean that data is treated differently each time it ‘crosses’ an invisible border – causing confusion and compliance hurdles for small businesses. We need a national privacy law with clear guidelines and standards for how companies collect, use, and share data. It would also help create transparency for consumers and ensure that companies can foster trust as the digital economy grows.
We also need a set of principles around responsible AI use. While I prefer less regulation, managing emerging technologies like AI requires clear rules that set out how this technology can be developed and deployed. With new innovations in AI unveiled almost daily, legislators should focus on guardrails and outcomes, rather than attempting to reign in specific technologies.
Rules should also consider the level of risk, focusing on AI systems that could meaningfully hurt Americans’ fundamental rights or access to critical services. As our government determines what ‘good policy’ looks like, industry will have a vital role to play. The Consumer Technology Association is working closely with industry and policymakers to develop unified principles for AI use.
We’re at a pivotal moment for the future of an amazing, complex and consequential technology. We can’t afford to let other countries take the lead.
Indian IT giant Infosys has launched a comprehensive and free AI certification training programme through Infosys Springboard to empower individuals with the necessary skills to succeed in the future job market.
Infosys Springboard offers a variety of AI courses, including introductory and advanced courses on topics like deep learning, natural language processing, and generative AI. Additionally, they provide a course called ‘Citizens Data Science’ which covers essential aspects of data science, such as Python programming, linear algebra, probability and statistics, and exploratory data analysis. This course aims to bridge the gap between theory and practice, enabling learners to acquire practical data science skills.
Infosys Springboard is a virtual platform offering a curriculum-rich learning experience accessible on any device. It fosters collaboration between educators and learners, catering to students from Class 6 to lifelong learners. With over 5.5 million registered users, it meets the increasing demand for AI education and positions Infosys as a trusted provider. Infosys Springboard aims to impact global education and career development.
Infosys’ Big Bets on Generative AI
Salil Parekh, CEO of Infosys, stated that the company’s operating margin for the entire year stands at 21%. He believes that Generative AI will create more opportunities and explained that the revenue was affected by cancellations and specific issues.
Back in 2015 when AI was not the hot topic, Infosys along with Elon Musk, AWS, and others donated $1 billion to OpenAI which eventually grew to be the world’s one of the most influential AI startups. Recently, Infosys co-founder Nandan Nilekani said that he will donate $38.5 million to his alma mater Indian Institute of Technology Bombay to push for AI advancements.
Infosys also recently launched Topaz, an AI-focused suite with 12,000 use cases, enabling industry-specific solutions in intelligent automation, AI-driven customer service, and enhanced security. Topaz helps businesses adopt open-source LLMs to build narrow transformers, solving specific enterprise challenges and driving growth.
Gary Bhattacharjee, VP of Data Strategy and AI at Infosys told AIM earlier that AI has the potential to eliminate entire functions rather than just improving processes, resulting in substantial cost reductions for clients. He highlighted the use of robotic process automation (RPA) at Infosys as an illustration of how AI can completely replace human decision-making processes. Bhattacharjee explained that RPA enables Infosys to achieve “labour arbitrage” by substituting human labour with low-cost robots.
The company is shifting its focus from coding to algorithms as coding becomes more automated. They aim to enhance their employees’ mathematical understanding and develop complex quant models. While this transformation will take time, Infosys will continue to hire freshers and collaborate with prestigious institutions like IITs and IIMs to attract talented individuals. They are also revamping their fresher program to prioritise skill development over coding.
However, Infosys has not sufficiently addressed the impact of AI adoption on the hiring of freshers. Although the company’s overall hiring, including both freshers and experienced software developers, has decreased from the previous fiscal year’s net addition of 380,000 to an estimated 280,000 in the current fiscal year (FY23), the effect on freshers remains uncertain.
As per a report by Goldman Sachs, AI could replace the equivalent of 300 million full-time jobs. So it is high time that you take up these courses to stay relevant.
The post Infosys Announces Free AI Training Program for Upskilling appeared first on Analytics India Magazine.
LinkedIn has already harnessed the writing abilities of generative AI with its AI-generated copy suggestions for ads on the platform. Now LinkedIn is expanding those abilities so that all users can take advantage of AI-generated copy.
LinkedIn posts are a great way for professionals to share the latest updates on their careers with their network. Typically, these posts include career moves, milestones, and accomplishments, which can help accelerate career growth.
Also: AI has the potential to automate 40% of the average work day
Carefully crafting what each post has to say can be a slightly challenging task, and LinkedIn's new feature is meant to be the solution, according to Keren Baruch, director of product at LinkedIn, who unveiled the feature on a LinkedIn post.
"When it comes to posting on LinkedIn, we've heard that you generally know what you want to say, but going from a great idea to a full-fledged post can be challenging and time-consuming," said Baruch. "So, we're starting to test a way for members to use generative AI directly within the LinkedIn share box."
To use the feature, users will have to provide the AI with at least 30 words outlining what exactly they want the post to convey. Then, AI will provide users with a first draft which they can then review and edit to post.
Also:How AI-generated content will revolutionize the ways we work
The post also shared a GIF of the feature in action, showcasing how easy the process of going from an idea to a fully crafted post on LinkedIn will be with this feature.
According to the post, the feature is in its testing stage before moving broadly to all users. However, this likely means that a select group of users will be accessing the feature.
There has been lots of backlash on the feature because people are arguing that it takes the authenticity away from LinkedIn posts. Another argument is that instead of interacting with humans on the app, it will be a bunch of chatbots interacting with each other.
Also:Six skills you need to become an AI prompt engineer
Although this is the first time AI post-generation would be native to the app, it is also important to note that users could be using external AI chatbots for assistance in creating posts already. So even without this feature, the authenticity of any posts you see may be questionable.
YouTube has been working towards expanding its content's accessibility for audiences of all languages. Most recently, in February, YouTube started testing multi-language audio, which would allow creators to dub their videos in 11 languages.
Now YouTube is turning to AI for its dubbing needs.
Also:Six skills you need to become an AI prompt engineer
As reported by The Verge, at VidCon, YouTube announced that it will be employing the use of Aloud, an AI-powered dubbing service from Area 120, Google's incubator for experimental products, to enable AI-powered dubbing for its videos.
The website for Aloud showcases how the dubbing process works. Aloud first transcribes the video. Then it can be reviewed and edited by the creator. Finally, the program translates and dubs the video so that it can be uploaded.
The site claims that transcriptions only take a few minutes and are available in Spanish and Portuguese. Although languages are limited at the moment, Google says it's planning on supporting Hindi, Bahasa-Indonesia, and other languages soon.
Also:Mass adoption of AI tools is derailing one very important factor, says MIT
Aloud is not yet released to the public, but there is a waitlist that users can join if they would like access. According to the report, YouTube is currently testing the feature with hundreds of creators.
TechCrunch+ roundup: AI + travel, fusion investor survey, why you’ll never get funding Walter Thompson @yourprotagonist / 9 hours
Most founders won’t realize that they’re on the wrong track until it’s too late.
That’s because failure is something many entrepreneurs can only perceive in hindsight: The day-to-day work of building a startup requires a high degree of confidence — and for some, denial.
Quoting the movie “Grosse Pointe Blank,” perhaps this column by Haje Jan Kamps will be “a swift, spiritual kick to the head that alters your reality forever.”
Full TechCrunch+ articles are only available to members Use discount code TCPLUSROUNDUP to save 20% off a one- or two-year subscription
In his experience, there are three reasons why some startups will never get a thumbs-up from an investor:
the market is too small;
the team is not good enough;
the plan doesn’t make sense.
These are all major problems, but here’s the good news: These are only existential issues for ego-driven founders. People who can accept they don’t have all the answers are able to pivot to success.
On the other hand, those who look in the mirror and see Captain America smiling back tend to make poor CEOs.
Have a fantastic weekend — go touch some grass.
Walter Thompson Editorial Manager, TechCrunch+ @yourprotagonist
Falling into these 3 traps means you will never raise VC funding
With $10T on the line, 6 fusion investors explain why they’re all in
Image Credits: Bryce Durbin / TechCrunch
Until December 2022, no one had achieved fusion ignition, the point at which the reaction produces more energy than is required to produce it.
“There’s still a long way to go, but net-positive controlled fusion is no longer just theoretical,” writes Tim De Chant.
To get an investor’s perspective on this nascent technology with a multitrillion-dollar TAM, he spoke to:
Katie Rae, CEO and managing partner, The Engine
Phil Larochelle, partner, Breakthrough Energy Ventures
Wal Van Lierop, founding partner, Chrysalix Venture Capital, and board member, General Fusion
Thai Nguyen, partner, MCJ Collective
With $10T on the line, 6 fusion investors explain why they’re all in
Get the TechCrunch+ Roundup newsletter in your inbox!
To receive the TechCrunch+ roundup as an email each Tuesday and Friday, scroll down to find the “sign up for newsletters” section on this page, select “TechCrunch+ Roundup,” enter your email, and click “subscribe.”
Click here to subscribe
Upgrading AI-powered travel products to first class
Image Credits: Westend61 (opens in a new window) / Getty Images
When it comes to integrating AI with travel, “even a small lead matters right now,” according to Eric Crowley and Adam Segall of global investment bank GP Bullhound.
Today’s consumers can create their own itineraries, but it still requires guesswork:
Can I walk to the beach from my hotel with a surfboard? Which restaurants near this conference center have vegan options?
In this TC+ post, Crowley and Segall share advice for founders working on AI-powered travel products: “At this early stage, our firm and other investors in the space we work with don’t have an expectation of perfection.”
Upgrading AI-powered travel products to first class
Why Europe and Israel’s unicorns are producing the next generation of tech founders
Image Credits: cerro photography (opens in a new window) / Getty Images
According to a report from global venture firm Accel, a “flywheel of inter-generational talent spawning from unicorns” is helping Europe and Israel maintain momentum even as deal flow and funding have slowed globally.
“Our data reveals that 221 of the region’s 353 VC-backed unicorns have fueled 1,171 new tech-enabled startups through their alumni,” writes Harry Nelis, a partner with the London-based company.
Why Europe and Israel’s unicorns are producing the next generation of tech founders
Warming public markets are boosting the secondary market for startup shares
Image Credits: skodonnell / Getty Images
Follow-on investment may be harder to obtain, but Alex Wilhelm reports that “the combination of seller pessimism and modest buyer optimism” is driving an “uptick in secondary market activity.”
Interpreting data from Caplight, Forge, Other Information and PitchBook, he concluded that the increase may signal a potential rise in startup valuations.
“It’s becoming clearer how much pain startups will need to endure to raise more capital, but it’s anybody’s guess if they’re willing to accept that fact.”
Warming public markets are boosting the secondary market for startup shares
Ask Sophie: Do I need 2 visas to work at 2 different startups?
Image Credits: Bryce Durbin/TechCrunch
Dear Sophie,
I’m in the U.S. on an H-1B visa, but I want to leave my current job and pursue a couple of startup ideas: One with a few friends, and the other on my own.
Do I need to get two separate visas to work at both companies at the same time? Can I transfer my H-1B to one or both companies?
— Energetic Entrepreneur
Ask Sophie: Do I need 2 visas to work at 2 different startups?
There’s a growing desire in the UK for more Black specialty venture funds
Row of white pencils and one black pencil, lying on a white background. Image Credits: Emilija Manevska / Getty Images
Black founders based in the United Kingdom have an even harder time than their U.S.-based counterparts when it comes to accessing venture capital, reports Dominic-Madori Davis.
“Between 2009 and 2019 in the U.K., just 30 Black people received VC funding, equating to less than 0.4% of all funds allocated to founders.”
She interviewed multiple investors who are working to expand this community, including Karl Lokko, co-founder of venture firm Black Seed, which recently announced its £5 million inaugural fund.
“The wheels are starting to turn, but we’ve got a long road ahead,” he said.
There’s a growing desire in the UK for more Black specialty venture funds
‘Get richer with ChatGPT’, ‘Making money online got easier’, ‘How you can make money with ChatGPT’, ‘Access my database of 125+ ChatGPT prompts to help you make money online for $5’.
Wait, stop, calm down!
So, if you’re fed up with all these paid advertisements trying to push different courses down your throats, luring you to make ‘easy’ money using ChatGPT, welcome to the club! Search engines and social media platforms like Instagram, Twitter, Facebook and even LinkedIn are brimming with such posts. These ‘influencers’ make ChatGPT seem like an easy money-minting machine.
But have people actually been able to monetise ChatGPT and its free version? Well, a lot of individuals claim to have been leveraging its capabilities to generate income. From branding to app-creation to providing writing services, they claim it opens avenues to earn ‘free money’ whether you’re an aspiring entrepreneur, content creator, or a freelancer.
You must’ve also come across tons of articles listing numerous ways to rake in the greens. These articles usually either bank on the chatbot’s capabilities to code and hint at it being used to build apps or websites ‘without any prior knowledge’ or for affiliate marketing, content marketing, optimising video production, or becoming a prompting expert amongst many other things.
However, there are some who claim to have made money this way. Ukrainian entrepreneur Ihor Stefurak built a Chrome extension using ChatGPT despite having zero coding knowledge. He claims to have generated $1000 in revenue within just 24 hours of launching the extension.
Many such claims and tutorials are flying fast and thick on YouTube, claiming to help make $100 a day but seem to know nothing about the tool, the model it is based on or even its name. Some coolly call it ChatGTP!
Meanwhile, there has also been a notable surge in AI-written ebooks on Amazon. These are ranked as high as 50,377 on the Kindle book store and also have five-star ratings in the hundreds. According to reports, as of mid-February there were over 200 e-books in Amazon’s Kindle store, listing ChatGPT as the author or co-author. The number may not seem like a lot but the actual number may be much higher since many authors do not disclose in the Kindle store that their book was written entirely, or in part, by a computer because Amazon’s policies do not require it.
In fact, the market is so saturated that making decent money out of this is improbable. According to Reuters, Brett Schickler, who wrote a children’s book titled, ‘The Wise Little Squirrel: A Tale of Saving and Investing’, (sold at the Amazon Kindle store for $2.99 and $9.99 for a printed version) netted less than $100.
Can Only Enhance
The reality is that ChatGPT alone cannot guarantee financial success. While ChatGPT can be an invaluable tool in reducing your workload, it is crucial to understand that it is not a magical solution that will automatically generate top-tier content for you. If you lack familiarity with a particular subject, you will struggle to differentiate between good and poor quality content.
For example, let’s consider a scenario where you follow someone’s advice on starting a successful blogging business with the assistance of ChatGPT despite having no prior experience in content creation. Even if ChatGPT generates a basic piece of content, you might perceive it as exceptional because you lack the knowledge of what constitutes top-tier content.
What you can actually do is create videos on how to make money with AI. As they say, ‘When there’s a gold rush, sell shovels and pickaxes’.
On a serious note, what you can actually do is automate parts of your workflow with ChatGPT APIs, which are free to use. One can take up email affiliate marketing, content marketing, or learn and preach better avenues like prompt engineering.
It is crucial not to be deceived by influencers who disseminate misleading content. Do not believe them when they claim that ChatGPT holds the secret to success or that it can effortlessly make you money. Instead, focus on learning and honing your skills, utilising ChatGPT as a tool to support you in accomplishing your goals.
The post Can ChatGPT Actually Make You Rich? appeared first on Analytics India Magazine.
With professional AI prompt engineering jobs going for $175,000 to well over $300,000 per year, prompt engineering is an attractive gig. But being a good AI prompt engineer involves more than being able to ask leading questions. You need to combine the disciplines of AI, programming, language, problem-solving, and even art to thrive on this career path.
Also: Do you like asking ChatGPT questions? You could get paid (a lot) for it
Prompt engineering is fundamentally the creation of interactions with generative AI tools. Those interactions may be conversational, as you've undoubtedly seen (and used) with ChatGPT. But they can also be programmatic, with prompts embedded in code, the rough equivalent of modern-day API calls; except, you're not simply calling a routine in a library, you're using a routine in a library to talk to a vast large language model.
Before we talk about specific skills that will prove useful in landing that prompt engineering gig, let's talk about one characteristic you'll need to make it all work: a willingness to learn. While AI has been with us for decades, the surge in demand for generative AI skills is new. The field is moving very quickly, with new breakthroughs, products, techniques, and approaches appearing constantly.
To keep up, you must be more than willing to learn — you must be voracious in learning, looking for, studying, and absorbing everything you possibly can find. If you keep up with your learning, then you'll be prepared to grow in this career.
Also: Want a job in AI? These are the skills you need
Here, then, are six skills we recommend you hone to become an AI prompt engineer.
1. Understand AI, ML, and NLP
A key place to start is building up an understanding of how artificial intelligence, machine learning and natural language processing actually work. If you're going to be interacting with large language models, you should understand what such a beast is, the different types of LLM out there, the types of things LLMs do well, and areas where they are weak.
This doesn't necessarily mean you need to become a computer scientist capable of creating your own LLM, but it does mean you need to understand a lot about the internals and capabilities of the tools you're trying to craft a career around. The key to this will be educating yourself by whatever means available, including traditional courseware, reading lots of articles and technical papers, attending conferences, and doing your own experiments.
One resource you should definitely check out is highlighted in this article by ZDNET's Sabrina Ortiz. She points you to a nine-part online class about using ChatGPT in development projects. The course is sponsored by OpenAI, the makers of ChatGPT and DeepLearning.ai, whose founder, Andrew Ng, teaches at Stanford and co-founded online learning giant Coursera. I'm taking the course and I recommend you do, too.
2. Define problem statements clearly and specify detailed queries
Fundamentally, this skill is the ability to communicate with clarity. Prompt engineering is all about how to tell the AI what you need. To do that, you need to get clear on what you want to get out of the interaction.
Also: How to write better ChatGPT prompts (and this applies to most other text-based AIs, too)
Here's an example. Let's assume you want to know more about Salem, the capital of Oregon. You need to be clear on at least two fronts. First, you need to explain the kinds of things you want to know, whether it's the political structure, issues of city management, traffic, or where the best donut shop is. Second, you need to be able to tell the AI that you're talking about Salem in Oregon, not the Salem in Connecticut, Virginia, or Indiana, or the witch trials in Salem in Massachusetts, or Winston-Salem in North Carolina, or any of the Salems in England, Wales, Australia, and Canada.
You'll also need to build up the skill of explaining how to set expectations for the AI, how to position it to understand the perspective it needs to use to provide value, and the context and scope of the problem you want it to solve in a given query.
Here, too, you'll need to understand the limits of various LLMs and how to work around them. For example, if you want a detailed white paper, you may need to first generate an outline, and then have the LLM write each section separately. Also, keep in mind that a clear prompt doesn't necessarily mean it's a short prompt. Longer prompts can result in more accurate and relevant responses.
The bottom line here is simple: Embrace clarity, and make sure you're able to communicate without making assumptions of understanding.
3. Be creative and develop your conversational skills
Prompt engineering is much more of a collaborative conversation than an exercise in programming. Although LLMs are certainly not sentient, they often communicate in a way that's similar to how you'd communicate with a co-worker or subordinate.
When you're defining your problem statements and queries, you will often have to think outside the box. The picture you have in your head may not translate to the internal representation of the AI. You'll need to be able to think about a variety of conversational approaches and different gambits to get the results you want.
Also: 6 things ChatGPT can't do (and another 20 it refuses to do)
Although I hope this isn't what you're going after, my best example of taking on conversational gambits is described in "How I tricked ChatGPT into telling me lies." My goal for that experiment was to get the AI to do something it was disinclined to do. Read through the article, and you'll see how I tried a number of creative approaches to find the conversational technique that yielded my desired results.
If you want to be a prompt engineer, experience on debate teams, negotiations, and even sales will stand you in good stead because they'll exercise those conversational, persuasion, and collaboration muscles that are so essential for eliciting desired results from generative AI systems.
4. Learn about writing and art styles, and build domain expertise
Not only will chatbots write answers for you, but they'll also often do it in the style you request. In "I used ChatGPT to rewrite my text in the style of Shakespeare, C3PO, and Harry Potter," I had more fun than any human has any right to have by asking ChatGPT to write things in the style of everything from Jane Austen to classic movie pirates. You haven't lived until you've read the preamble of the US Constitution written by a pirate!
Also: How I used ChatGPT and AI art tools to launch my Etsy business fast
Those examples were purely for fun and experimentation, but I also used the "write in the style of" preface for setting up my experimental Etsy store. I had ChatGPT write copy in the style of Jony Ive, whose excessively flowery descriptions of Apple products have become the stuff of legend.
You can use styles for graphics generators like Midjourney as well. You can draft images in cinematic style, in the style of 1940s cartoons, and in a wide range of photographers and art styles.
For example, I used my standard Facebook icon picture and fed it to Midjourney with the prompt "cubism" and this is what resulted.
This is the original image I fed into Midjourney. As you'll see, it gets a bit confused by the brass microphone, but it's still interesting.
In this case, I kept the prompt minimal, but used what's called a "seed image," which was my Facebook icon. Here are four variations using the "cubism" style. Which do you think is most like the original?
In addition to understanding writing and art styles, it's important for you to develop (or be able to access) the domain expertise of the area you're setting up prompts for. For example, if you're working on an AI application for auto diagnostics, it's important for you to have enough familiarity to be able to elicit the responses you need and understand if they're correct or wrong.
Also: The best AI art generators: DALL-E 2 and other fun alternatives to try
Oh and here's a skill within the skill: learn about the extensions and plugins that your favorite generative AI tools use. As time goes on, those extensions and plugins will help you do things you can't do with the off-the-shelf AI tool. So learning about and using add-ons will not only keep your skills fresh, but will let you accomplish things not otherwise possible.
5. Develop scripting and programming skills
Did you ever notice that whenever someone prefaces a phrase with "it goes without saying," there's gonna be some saying happening? In any case, it goes without saying (but I'm going to say it) that programming skills would come in handy. While there will be some prompt engineering gigs that interact merely with the chatbots, the better-paying gigs will likely involve embedding AI prompts into applications and software that then provide unique value.
Also: How to use ChatGPT to create an app
While you might not necessarily be expected to write the full application code, you will provide far more value if you can write some code, test your prompts in the context of the apps you're building, run debug code, and overall be part of the interactive programming process. It will be much easier for a team to move forward if the prompt engineering occurs as an integral part of the process, rather than having to add it in and test it as a completely separate operation.
And, besides, coding skills are valuable all on their own. Everyone should have some basic coding experience.
6. Build up your patience (and sense of humor)
I'm a firm believer that it's much easier to be patient if you have a sense of humor. Something that's infuriating can be less toxic to your soul if you can see the essential humor in that annoyance. These generative AIs definitely require patience. They'll completely misinterpret requests. They'll lose the thread of a conversation right when you're about to have a breakthrough. They'll completely fabricate answers that are total BS.
If you can't chuckle about some of it, you're destined to have a rough time.
That's the case with programming, as well. Every programmer needs patience. One of the biggest challenges some of my students had when starting out programming was that they couldn't accept that their code wouldn't work the first time it ran. Those who couldn't stick it out and do the work didn't complete the course. By contrast, even those who were less than inspired coders, but had the patience to try, fail, research some more, and try again were very successful.
Also: The most popular programming languages and where to learn them
Think of it this way. AI prompting is a mix of working with an incredibly literal computer, a willful learning model that interprets things in unpredictable ways, human team members (some of whom are even more literal than the machines), and the randomly unpredictable nature of the universe.
Patience isn't just a virtue. It's a superpower.
Some additional words of wisdom
So, there you go. I've outlined six skills/characteristics you need to find success as a prompt engineer. But keep in mind that two paragraphs saying "learn about AI" isn't going to get you there. These are just rough guidelines, and it's a very individualized path ahead of you that you'll need to follow.
If I can give you one essential piece of advice, it's this: Tinker. Pick projects of your own that interest you and build something. Team up with a few friends and see what you can produce. Having some hands-on experience will take you a lot farther than a list from some guy on the Internet.
Also: 73% of consumers trust what generative AI wants us to see
Get out there and do some prompt engineering. Build some small applications. Take that course I recommended. Build some stuff. Then, not only will you no longer be someone who wants to go into prompt engineering, you'll be someone who's done it and has something to show for it.
Do it. If you do it, you will be it.
Disclaimer: Using AI-generated images could lead to copyright violations, so people should be cautious if they're using the images for commercial purposes.
You can follow my day-to-day project updates on social media. Be sure to follow me on Twitter at @DavidGewirtz, on Facebook at Facebook.com/DavidGewirtz, on Instagram at Instagram.com/DavidGewirtz, and on YouTube at YouTube.com/DavidGewirtzTV.
EvaluAgent raises $20M to build out software that evaluates call center agents Kyle Wiggers 10 hours
After spending more than a decade in the customer experience industry, three friends, Jaime Scott, Michelle Dinsmore and Alex Richards, decided to launch their own company, EvaluAgent, to develop quality assurance testing software for contact centers.
The three co-founders say they were motivated by a shared desire to find a solution to the problem of contact center manager overwork. Reviewing customer calls and texts for quality assurance calls takes time — so much time, in fact, that it’s rarely done regularly. According to one source, only between 1% and 2% of calls to a contact center actually get evaluated.
“At its heart, the contact center’s role is to serve customers,” Scott, who serves as the company’s CEO, told TechCrunch in an email interview. “Those customers are human beings and, as such, are always going to demand a level of service and experience technology alone won’t be able to offer. Our belief is that the greater the employee experience for agents, the greater the customer experience they’ll be able to offer.”
Scott, Dinsmore and Richards incorporated EvaluAgent in 2012. The team grew quickly, but was largely focused on working with a small number of corporate clients. That changed in 2018 and 2019, when Scott says the leadership recognized a gap in the market for a more flexible software-as-a-service-based quality assurance testing solution.
Today, EvaluAgent’s platform aims to help quality assurance staff analyze conversations — both text and voice — across channels to coach and train customer agents. Via mostly automated workflows, EvaluAgent tries to boost the efficiency of QA teams, showing quality assurance-related stats in a unified dashboard.
Managers can give employee feedback via EvaluAgent’s dedicated tool while employees can find answers to common questions in EvaluAgent’s built-in company knowledge base.
“EvaluAgent not only delivers quality assurance, but also provides customers with tools to ensure that evaluators’ findings convert to ongoing behavior change in the agent base,” Scott said. “Altogether, the platform represents a complete system of record for contact centers’ quality teams.”
EvaluAgent also offers an automated scoring system, driven by a combination of speech recognition technologies and OpenAI’s ChatGPT. The system, SmartScore, transcribes conversations if necessary and uses ChatGPT — an AI chatbot that understands text — to summarize them for insights, rating line items and providing coaching tips.
“Out-of-the-box key moments, such as customer frustration, repeat contact, and complaint escalation, are automatically tagged,” Scott explained. “It … empowers organizations to achieve more interaction coverage and streamline the quality assurance process while keeping quality teams in full control.”
Image Credits: EvaluAgent
One wonders about bias creeping into the algorithms used to evaluate agents’ actions, though. After all, studies have shown that AI is more likely to classify Black speech as “toxic” or “offensive.” And it’s well-understood that voice recognition tech, too, is racially and ethnically biased. So can SmartScore be trusted?
Scott didn’t elaborate on which measures, if any, EvaluAgent takes to combat bias in its algorithms. But he stressed that they aren’t intended to replace human quality assurance evaluators.
“EvaluAgent has made investments in its technical resources to incorporate the best AI models available for its use cases,” he said. “We are model-agnostic, which we believe is a significant advantage in the modern rapidly-evolving AI landscape. Additional AI-fueled innovation and automation, which incorporate the latest AI models but allow quality assurance teams to remain at the center of key workflows, will follow on the back of this financing.”
The lack of transparency hasn’t gotten in the way of the company’s success, it seems. EvaluAgent claims it’s seen revenue grow nearly fivefold over the last three years, with customers ranging from Fortune 500 enterprises to mid-market business process outsourcers and tech companies.
The momentum captured the attention of major investors, which poured cash into EvaluAgent’s Series A. Announced today, the Series A, led by PeakSpan, closed at $20 million, bringing EvaluAgent’s total raised to $21 million.
PeakSpan’s Phil Dur, the lead partner on the deal, says that he saw “tremendous opportunity” in what he described as the “quality arena.” That’s despite stiff competition in the over-$35.3-billion space, apparently — startups including Invoca offer features similar to — but not exactly the same as — EvaluAgent’s.
“Having spent years studying the category, we believe EvaluAgent is the strongest vendor of quality assurance and performance improvement software,” Dur said via email. “The platform shines for small and mid-market contact centers, while maintaining the capacity to serve enterprise-level businesses with large contact center operations.”
Scott says that the Series A funding will be put toward enriching its products, expanding its remote team and supporting customers in “new and existing geographies.”
“Roughly 25% of the contact center market still runs their QA programs on spreadsheets, while another portion makes the best of in-house legacy tools,” he said. “Contact centers are facing a new dynamic — balance leaner staff and fewer onsite agents with an increasingly demanding customer (call volumes, hold times and escalations have been steadily rising) all while optimizing cost to navigate a potentially choppy macroenvironment.”
Amazon Web Services (AWS) has introduced the AWS Generative AI Innovation Center, a program designed to assist customers in the development and implementation of generative artificial intelligence (AI) solutions. With a $100 million investment, AWS will connect its AI and machine learning (ML) experts with customers worldwide, enabling them to design, and launch new generative AI products, services, and processes. This initiative builds upon AWS’s extensive experience in AI technology and forms part of their overarching strategy to democratise generative AI on a global scale.
The AWS Generative AI Innovation Center aims to help these companies turn their ideas into reality more efficiently. The centre’s team, comprising strategists, data scientists, engineers, and solutions architects, will collaborate with customers to develop customised solutions that harness the power of generative AI. Examples include accelerating drug research and discovery in healthcare, reinventing industrial design and processes in manufacturing, and delivering personalised information and advice in the financial services sector.
Through no-cost workshops, engagements, and training, AWS will guide customers in identifying use cases that create the most value for their businesses, based on industry expertise and best practices. Customers will closely collaborate with generative AI experts from AWS and the AWS Partner Network to select appropriate models, overcome technical or business challenges, develop proofs of concept, and plan for large-scale deployment. The Generative AI Innovation Center team will also provide guidance on responsible generative AI practices and optimising machine learning operations for cost reduction.
Engagements with customers will involve the provision of strategies, tools, and assistance in utilising AWS generative AI services. These services include Amazon CodeWhisperer, an AI-powered coding companion, and Amazon Bedrock, a managed service that offers foundational models from various providers, including AI21 Labs, Anthropic, Stability AI, and Amazon’s own family of models called Amazon Titan.
Customers can train and execute their models using high-performance infrastructure such as AWS Inferentia-powered Amazon EC2 Inf1 Instances, AWS Trainium-powered Amazon EC2 Trn1 Instances, and Amazon EC2 P5 instances powered by NVIDIA H100 Tensor Core GPUs. Moreover, customers have the option to build, train, and deploy their own models using Amazon SageMaker or utilise Amazon SageMaker Jumpstart to deploy popular models like Cohere’s large language models, Technology Innovation Institute’s Falcon 40B, and Hugging Face’s BLOOM.
Highspot, a sales-enablement platform, leverages the AWS Generative AI Innovation Center to transform sales enablement and deliver enhanced value to customers. Lonely Planet, a travel company, aims to provide personalised travel recommendations by utilising the centre in conjunction with expert-driven advice and their award-winning content. Twilio, a customer engagement platform, seeks to empower businesses with both generative and predictive intelligence capabilities to better understand and provide value to their customers.
The AWS Generative AI Innovation Centre’s mission is to foster new ideas and capabilities in the field of AI, enabling organisations to leverage generative AI for their benefit.
The post AWS Announces Generative AI Innovation Centre appeared first on Analytics India Magazine.
Authors: Michael Ortega and Geoffrey Angus Make sure to register for our upcoming webinar to learn how to use large language models to extract insights from unstructured documents. Thanks to ChatGPT, chat interfaces are how most users have interacted with LLMs. While this is fast, intuitive, and fun for a wide range of generative use cases (e.g. ChatGPT write me a joke about how many engineers it takes to write a blog), there are fundamental limitations to this interface that keep them from going into production.
Slow — chat interfaces are optimized to provide a low-latency experience. Such optimizations often come at the expense of throughput, making them unviable for large-scale analytics use cases.
Imprecise — even after days of dedicated prompt iteration, LLMs are often prone to providing verbose responses to simple questions. While such responses are sometimes more human-intelligible in chat-like interactions, they are oftentimes more difficult to parse and consume in broader software ecosystems.
Limited support for analytics- even when connected to your private data (via an embedding index or otherwise), most LLMs deployed for chat simply cannot ingest all of the context required for many classes of questions typically asked by data analysts.
The reality is that many of these LLM-powered search and Q&A systems are not optimized for large-scale production-grade analytics use cases.
The right approach: Generate structured insights from unstructured data with LLMs
Imagine you’re a portfolio manager with a large number of financial documents. You want to ask the following question, “Of these 10 prospective investments, provide the highest revenue achieved by each company between the years 2000 to 2023?” An LLM out-of-the-box, even with an index retrieval system connected to your private data, would struggle to answer this question due to the volume of context required.
Fortunately, there’s a better way. You can answer questions over your entire corpus faster by first using an LLM to convert your unstructured documents into structured tables via a single large batch job. Using this approach, the financial institution from our hypothetical above could generate structured data in a table from a large set of financial PDFs using a defined schema. Then, quickly produce key statistics on their portfolio in ways that a chat-based LLM would struggle.
Even further, you could build net-new tabular ML models on top of the derived structured data for downstream data science tasks (e.g. based on these 10 risk factors which company is most likely to default). This smaller, task-specific ML model using the derived structured data would perform better and cost less to run compared to a chat-based LLM.
Learn how to extract structured insights from your documents with LLMs
Want to learn how to put this approach into practice using state-of-the-art AI tools designed for developers? Join our upcoming webinar and live demo to learn how to:
Define a schema of data to extract from a large corpus of PDFs
Customize and use open-source LLMs to construct new tables with source citations
Visualize and run predictive analytics on your extracted data
You’ll have a chance to ask your questions live during our Q&A.
Save your spot
More On This Topic
Unstructured Data: The Must-Have For Analytics In 2022
A Structured Approach To Building a Machine Learning Model
Building a Structured Financial Newsfeed Using Python, SpaCy and Streamlit
Exploratory Data Analysis Techniques for Unstructured Data
8 Free AI and LLMs Playgrounds
What are Vector Databases and Why Are They Important for LLMs?