How AI might threaten democracy, plus OpenAI making AI phone?

The December 27 intelligst AI newsletter

Author

Joaquin Bueno
December 27, 2023

Greetings, intelligists!

Hopefully, yesterday’s doomsday newsletter didn’t scare any of you off. I did feel the need to spread some important perspectives on what unbridled AI development could imply.

Here at intelligist, we believe in the positive and transformative impacts that AI can have on society. I don’t doubt for a second that AI-driven technologies will improve many lives. I, for one, harbor great hope for a future where AI can be used for improving medicine, reducing meaningless work, lowering environmental devastation, and advancing humankind.

But this optimism should not overlook the potential downsides. There are definitely dangers, as Raskin and Harris point out. My hope is that we see them, and not only in hindsight, when it might be too late.

That being said, here’s today’s digest. Enjoy what’s left of 2023!

Joaquin

OpenAI device coming?

Apple has just been dealt a major blow, while AI mavericks OpenAI have won out big-time.

Apple's long-time iPhone and Watch design chief, Tang Tan, has decided to join design legend Jony Ive’s design team. They will be working with Sam Altman at OpenAI, and the rumors hint that they will attempt to develop a proprietary AI device. Some speculate this could be the dawn of an "iPhone of artificial intelligence."

As Apple's design team witnesses a reshuffle, the industry eagerly anticipates what this collaboration could bring to the AI landscape. Apple has notoriously lagged behind other companies in implementing AI in its devices.

2024 Election: Navigating the AI Misinformation Tsunami

With the 2024 U.S. Presidential election on the horizon, experts are sounding the alarm on the potential onslaught of AI-driven misinformation.

Previous elections have already seen the significant impact of foreign actors using social media to influence voters.

However, AI tools are making it easier than ever to spread disinformation (especially in the form of fake videos, images, texts, and more). This creates massive opportunities to malicious foreign actors who might gain from influencing the vote.

Whistleblowers warn that the impending 'digital storm' could challenge the foundations of democracy, as deepfakes and reduced social media vigilance make finding the truth increasingly difficult. With the current state of the world, balanced on a dangerous precipice of widespread war, there are certainly reasons to be concerned.

And while there are many concerns, there are very few solutions. Right now, there are no mechanisms for large-scale AI regulations. Implementing any would take time, effort, and bipartisan cooperation in the US. At the same time, there are no universally reliable tools for discerning real images or videos from fakes. Overall, the current trend is for such tools to become more widespread.

While some companies, such as OpenAI, have taken measures, there are ways around them. For example, OpenAI prevents users from generating images of famous people. However, there are open-source platforms, such as Stable Diffusion, which have no censorship controls for users, who can install them on their own devices and modify them.

This is a double-edged sword, as open-source is considered by many experts and advocates to be a crucial counterbalance to corporate software. Yet, open-source tools are precisely what many election hackers will use to attempt to influence the vote.

Microsoft’s Copilot AI now on Android

Microsoft’s Copilot AI assistant, powered by OpenAI's latest models, is now available on Android.

This strategic expansion to mobile platforms signals Microsoft's commitment to becoming the world’s dominant force in AI. Their aim is to make AI more accessible, with Copilot's versatile capabilities ranging from coding assistance to creative image generation.

This is also a powerful move from Microsoft, who undoubtedly are attempting to gain ground on Apple and other technological rivals.

Read More: Copilot AI Lands on Android​​

Axios: 2024 will truly be the year of AI

If you thought 2023 was a big year for AI, you might be in store for a surprise.

Analysts at Axios have stated that AI will “get real” in 2024. They noted that AI's journey in 2024 is set to shift from hype to tangible impact. Change wil affect virtually all sectors of the economy and society.

Key areas include education, the workplace, where AI co-pilots could redefine roles, and healthcare, where AI promises earlier disease detection and personalized treatments.

Amidst this, the industry faces significant challenges like managing hallucinating chatbots, addressing copyright issues, potential AI dangers, and navigating regulatory landscapes. 

So, if you thought there was an AI explosion in 2023, the coming year is poised to be even bigger. Many technologies that emerged this year could soon become standard tools around the world.