Subscriber count closing in on 1400!

We crossed 1000 on April 16th. I love the growth but realize that so many people might find value in this that don’t even know about it. Now listen, I have the best readers. You all have been sharing this newsletter graciously. I am humbled.

So, I welcome your ideas to how to grow it more. I have the opportunity to take on ads (not much $ at all). I could use that $ to buy ads or promote it on Twitter or LinkedIn. But I really like organic growth rather than paid growth.

So with you big Brainyact brains, I ask you - what ideas might you have?

Ok, time to dig in!

A special Hello 👋 to my NEW SUBSCRIBERS! 
To read previous posts, go here.

Today we will:

1. shiver as winter is coming

2. actually use Bing Chat for a use case and prompt

3. share 4 tools

4. chuckle at OpenAI trying to Super Bowl™️

5. read a joint statement on AI from 4 US agencies

6. get a new free Chat tool

7. learn what “attack surface area” means

8. cringe at AI created political ads

9. enjoy a meme

Winter Is Coming

The Impending Generative AI Winter: A Call for Insight and Prudence

As generative AI continues to experience rapid advancements in areas such as large language models, image synthesis, and audio generation, it is essential to consider the potential challenges and limitations that lie ahead. The forthcoming "generative AI winter" could be a consequence of not only the ethical, social, and technical issues that plague the field but also the result of an unsustainable hype cycle that drives unrealistic expectations.

What is a “generative ai winter?”

A "generative AI winter" in plain language refers to a period of stagnation or decline in the development, interest, or progress of generative artificial intelligence technologies. During this time, researchers might face difficulties in making significant breakthroughs, funding for AI projects may decrease, and overall enthusiasm for generative AI may wane. This can be due to various factors, such as technological limitations, unrealistic expectations, or external economic factors. The term "winter" is used metaphorically to signify a slow or challenging phase, much like the cold and barren season.

I believe we are heading for a winter by the end of the year for four main reasons detailed below.

1. The Dark Side of Generative AI: Ethical and Social Challenges

Generative AI's impressive capabilities have given rise to significant ethical and social concerns. Malicious uses of this technology include the spread of misinformation, creation of deepfakes, and impersonation. Additionally, questions about intellectual property, privacy, and human creativity are increasingly pertinent as AI-generated content becomes more prevalent.

The danger of overlooking these concerns is that they could contribute to the eventual disillusionment with generative AI technologies. If not adequately addressed, these issues could lead to a backlash against the technology, damaging its reputation and hindering future progress.

Regulators are piling in now and this will cause confusion, jurisdictional difference, and much hand wringing - all of which will gum up the works of AI advances.

See:
Rand Report: Artificial Intelligence, Deepfakes, and Disinformation
PIA: Responsible Practices for Synthetic Media

2. The Unsustainable Hype Cycle: A Glut of Startups and Overpromising

The current generative AI hype cycle may not be sustainable in the long term, as many startups and investors rush to capitalize on the potential applications and markets. Unfortunately, not all of these startups have viable business models, clear value propositions, or differentiated products. Some may even be overpromising or underdelivering on generative AI's capabilities.

The likely outcome of this unsustainable growth is a loss of confidence and interest in generative AI from investors and consumers, as the competition increases and companies fail to survive or scale. This disillusionment could lead to a generative AI winter, characterized by reduced investment, stalling innovation, and diminished public interest in the field.

The AI Hall of Shame will no doubt fill its halls again.

See:
AI, generative AI boom at Y Combinator Demo Day
ChatGPT Has Investors Drooling—but Can It Bring Home the Bacon?
2023 data, ML and AI landscape: ChatGPT, generative AI and more

3. The Democratization and Subsequent Corporatization of Generative AI

The recent accessibility of generative AI to the general public, exemplified by the release of ChatGPT's free model, has sparked euphoria and mass adoption. However, as the domain matures, we are witnessing an increasing number of corporations developing proprietary models, which in turn puts access behind paywalls and limits transparency into controlled datasets and learning models.

While these corporations may make it easier for users to extract value from generative AI by simplifying its usage or integrating it into existing products, the potency and free reign of raw access will gradually diminish. This trend could stifle innovation and concentrate the power of generative AI in the hands of a few major players, leading to a less diverse and open ecosystem.

See:
Democratization of AI creates benefits and challenges
The Rise of Generative AI and the Democratization of Data

4. The Open-Source Paradox: Greater Access, Yet Limited Adoption

The growth of open-source models offers an alternative to the corporatization of generative AI. However, the technical proficiency required to use these models effectively can be a barrier for casual users who lack the necessary skills or knowledge. As a result, the democratization of generative AI becomes limited, as many individuals who might have benefited from direct access to the technology are unable to leverage it.

For instance, while posting a project on GitHub could be an excellent way to share and collaborate on generative AI models, the majority of regular people have no idea what GitHub is or how to use it. This disconnect creates a gap between the potential of open-source generative AI and the actual adoption among the wider population.

See:
On the emerging landscape of open AI

Striding Into a Chilling Season: Our Collective Opportunity

The fast-approaching generative AI winter is not a fait accompli but rather a call to arms for researchers, policymakers, and users alike. We must collectively embrace the challenges and harness the opportunities that lie ahead.

In this chilling season, let us remember that the frost serves as a reminder to prepare, not to despair. The time has come to rethink our approach to generative AI, ensuring that we pave a path that leads to a vibrant spring, bursting with innovation, inclusivity, and responsible growth.

Use Case Just Discovered

Oh my gosh! I am about to give you a Bing Chat use case and prompt!

It’s rare that I use Bing Chat but I have consciously tested it more and more. Today I had a specific need and BAM! it delivered just minutes before I am writing this.

I wrote the above essay Winter is Coming based on my own thoughts but also reviewing a number of resources. In writing it I found the original resources were inspirations but that I had departed greatly from what they covered.

At first, I was thinking about just shipping the essay without resources but then I had an unnerving feeling. You the reader might ask yourself, who the heck does Josh think he is to just declare a winter is coming? So, I quickly tried to uncover the tracks of my research. It was frustrating, time-consuming, and returning poor results. And that is when it hit me . . .

Use Bing Chat! It is connected to the damn internet and might be able to find you some resources based on my original ideas that I wrote about. So that is what I did.

Now, this is a simple prompt but as I always say, we need better use cases not prompts. We need better more clever ways to use conversational ai, not a book of 1000 prompts!

▶︎▶︎PROMPT

Could you review the following article I am writing to find possible sources to cite for the main themes? [paste the article]

Now Bing Chat is no ChatGPT in terms of its robustness or comprehensivness. I find it is strong at these discrete tasks. But as you will see you have to feed it slowly.

It only analyzed half of the essay.

▼▼RESPONSE

Here is a link to it if you have Microsoft Edge downloaded.

I added the parts of my essay it ignored to get a list of resources. I reviewed the resources and indeed they support the main points in each section for the most part.

So the next time you are writing something and want some resources to back you up or to reference in refining your thoughts, pop your draft into Bing Chat and ask it to find some resources. Plus I am sure I can finely tune this prompt to be even better at this task

Misc Tools & Resources: 

Pace - Generate ideas in seconds instead of days with the power of AI

Flash Insights - Effortlessly extract valuable insights from any webpage or video in seconds

MyQueue - Save articles you want to read and listen to on the go

Sumly.ai - AI-generated podcast summaries delivered straight to your inbox.

News you can Use: 

OpenAI wants to “Super Bowl™️” GPT

In a race against AI-obsessed rivals, OpenAI tried to pull a fast one on the U.S. Patent and Trademark Office, hoping to secure the "GPT" trademark in record time. But alas, the office wasn't having any of it, leaving OpenAI to cool their jets alongside the rest of the trademark-seeking mortals.

It seems OpenAI's got ants in their pants after a slew of GPT copycats emerged, including a chatty upstart from Elon Musk himself.

EEOC, FTC, and Other Federal Agencies Release Joint Statement on Confronting Bias and Discrimination in AI and Automated Systems

🆓 A new FREE Chat tool was released.

https://huggingface.co/chat/

Hugging Face has launched HuggingChat, an open-source alternative to ChatGPT that allows users to interact with Open Assistant, an open-source chat assistant organized by nonprofit LAION. The platform aims to increase transparency, inclusivity, accountability, and distribution of power in AI, and has drawn comparisons to the Android App Store. However, HuggingChat currently faces limitations, including licensing issues that may prevent commercial use and being in its initial version.

Pros:

1. Open-source alternative to ChatGPT, promoting transparency and inclusivity

2. Increases accountability and distribution of power in AI

3. Potential to become the equivalent of the Android App Store

Cons:

1. Licensing issues may prevent commercial use

2. Currently in its initial version, with many limitations and a need for further iteration on interface and safety mechanisms

☢️ 🔫 Enterprise AI increases the “attack surface area” of your organization.

An attack surface area is the entire area of an organization or system that is susceptible to hacking. It's made up of all the points of access that an unauthorized person could use to enter the system and cause damage by manipulating or downloading data. An attack surface area includes both the digital and physical aspects of the system, such as hardware, software, applications, code, ports, servers, websites, devices, and users. An attack surface area can be reduced by applying security best practices, such as encryption, authentication, authorization, monitoring, auditing, testing, and threat modeling.

News you can Lose: Political ads go AI. Please! No!!

In the Meme Time:

That's a wrap for today. Stay thirsty & see ya next time! If you want more, be sure to follow me on Twitter and LinkedIn.

DISCLAIMER: None of this is legal advice. This newsletter is strictly educational and is not legal advice or a solicitation to buy or sell any assets or to make any legal decisions. Please /be careful and do your own research.