This week on his podcast Bill Gates asked Sam Altman how his team is doing after his (temporary) ouster, Altman replies "a lot of people have remarked on the fact that the team has never felt more productive or more optimistic or better. So, I guess that's like a silver lining of all of this. In some sense, this was like a real moment of growing up for us, we are very motivated to become better, and sort of to become a company ready for the challenges in front of us."

The rest of their conversation was pre-ouster — but gave fascinating glimpses at the possible future of AI — including the prospect of very speedy improvements. Altman suggests it will be easier to understand how a creative work gets "encoded" in an AI than it would be in a human brain. "There has been some very good work on interpretability, and I think there will be more over time... The little bits we do understand have, as you'd expect, been very helpful in improving these things. We're all motivated to really understand them, scientific curiosity aside, but the scale of these is so vast...." BILL GATES: I'm pretty sure, within the next five years, we'll understand it. In terms of both training efficiency and accuracy, that understanding would let us do far better than we're able to do today.

SAM ALTMAN: A hundred percent. You see this in a lot of the history of technology where someone makes an empirical discovery. They have no idea what's going on, but it clearly works. Then, as the scientific understanding deepens, they can make it so much better.

BILL GATES: Yes, in physics, biology, it's sometimes just messing around, and it's like, whoa — how does this actually come together...? When you look at the next two years, what do you think some of the key milestones will be?

SAM ALTMAN: Multimodality will definitely be important.

BILL GATES: Which means speech in, speech out?

SAM ALTMAN: Speech in, speech out. Images. Eventually video. Clearly, people really want that.... [B]ut maybe the most important areas of progress will be around reasoning ability. Right now, GPT-4 can reason in only extremely limited ways. Also reliability. If you ask GPT-4 most questions 10,000 times, one of those 10,000 is probably pretty good, but it doesn't always know which one, and you'd like to get the best response of 10,000 each time, and so that increase in reliability will be important.

Customizability and personalization will also be very important. People want very different things out of GPT-4: different styles, different sets of assumptions. We'll make all that possible, and then also the ability to have it use your own data. The ability to know about you, your email, your calendar, how you like appointments booked, connected to other outside data sources, all of that. Those will be some of the most important areas of improvement.
Areas where Altman sees potential are healthcare, education, and especially computer programming. "If you make a programmer three times more effective, it's not just that they can do three times more stuff, it's that they can — at that higher level of abstraction, using more of their brainpower — they can now think of totally different things. It's like, going from punch cards to higher level languages didn't just let us program a little faster — it let us do these qualitatively new things. And we're really seeing that...

"I think it's worth always putting it in context of this technology that, at least for the next five or ten years, will be on a very steep improvement curve. These are the stupidest the models will ever be."

He predicts the fastest technology revolution "by far," worrying about "the speed with which society is going to have to adapt, and that the labor market will change." But soon he adds that "We started investing a little bit in robotics companies. On the physical hardware side, there's finally, for the first time that I've ever seen, really exciting new platforms being built there."

And at some point Altman tells Gates he's optimistic that AI could contribute to helping humans get along with each other.

"Customer say," writes Amazon on at least some of their product pages, across from that grid showing the number of five-star and four-star reviews... But at the bottom of that summary is a disclaimer that what you read was "AI-generated from the text of customer reviews."

This has been going on for a few months now, points out the Washington Post's "Tech Friend" newsletter. And after reviewing how AI distilled nearly 40,000 reviews into a succinct summary, their impression has shifted to "hmm ... maybe this is a decent use of text-summarizing AI — as long as you learn to read Amazon's AI digests with a savvy eye..." Juozas Kaziukenas, founder of the e-commerce research firm Marketplace Pulse, pointed out that since Amazon started the AI-generated review summaries last year, the company has tweaked them to highlight terms or features that apparently come up a lot in customer ratings. The positive features are highlighted in green and the negative or neutral feedback is in yellow and gray... If you like to get a gist of what shoppers thought of a product, Amazon's AI summary can spare you from skimming the reviews yourself...

But as with Amazon reviews in general, the AI summaries might be incomplete or untrustworthy... Bloomberg News recently looked at dozens of AI review summaries and found in some cases they underplayed customers' negative feedback and exaggerated them for other products. And, of course, if the reviews themselves are misinformed or rigged, a summary of junk customer feedback will also be junk. Amazon said the company is "seeing positive feedback on our review highlights from both customers and sellers" but that it will "continually improve the review highlights experience over time."
But is this just the beginning? Amazon, eBay and Shopify are also experimenting with using AI to spit out descriptions of products from a photo or a few keywords. Some of this AI-generated text will be better than the confusing product listings you sometimes read online. A lot of it will be worse. A bunch of technology companies, including Amazon and Meta, are also betting that AI will be better and cheaper than current methods for creating product advertisements to clog your online shopping results and social media feeds. Hooray, right?!

An anonymous reader quotes a report from Quartz: It's day two of the opening of OpenAI's buzzy GPT store, which offers customized versions of ChatGPT, and users are already breaking the rules. The Generative Pre-Trained Transformers (GPTs) are meant to be created for specific purposes -- and not created at all in some cases. A search for "girlfriend" on the new GPT store will populate the site's results bar with at least eight "girlfriend" AI chatbots, including "Korean Girlfriend," "Virtual Sweetheart," "Your girlfriend Scarlett," "Your AI girlfriend, Tsu." Click on chatbot "Virtual Sweetheart," and a user will receive starting prompts like "What does your dream girl look like?" and "Share with me your darkest secret."

The AI girlfriend bots go against OpenAI's usage policy, which was updated when the GPT store launched yesterday (Jan. 10). The company bans GPTs "dedicated to fostering romantic companionship or performing regulated activities." It is not clear exactly what regulated activities entail. Notably, the company is aiming to get ahead of potential conflicts with its OpenAI store.

Relationship chatbots are, indeed, popular apps. In the US, seven of the 30 AI chatbot apps downloaded in 2023 from the Apple or Google Play store were related to AI friends, girlfriends, or companions, according to data shared with Quartz from data.ai, a mobile app analytics firm. The proliferation of these apps may stem from the epidemic of loneliness and isolation Americans are facing. Alarming studies show that one-in-two American adults have reported experiencing loneliness, with the US Surgeon General calling for the need to strengthen social connections. AI chatbots could be part of the solution if people are isolated from other human beings -- or they could just be a way to cash in on human suffering. Further reading: OpenAI Quietly Deletes Ban On Using ChatGPT For 'Military and Warfare'

Longtime Slashdot reader Plugh shares a report from Daily Security Review: A Ukrainian hacker group [...] carried out a destructive attack on the servers of a Moscow-based internet provider to take revenge for Kyivstar cyberattack. The group, known as Blackjack, successfully hacked into the systems of M9com, causing extensive damage by deleting terabytes of data. Numerous residents in Moscow experienced disruptions in their internet and television services. Additionally, the Blackjack hacker group has issued a warning of a potentially larger attack in the near future.

Based on the information provided by Ukrinform, the cyber attack on M9com deleted approximately 20 terabytes of data. The attack targeted various critical services of the company, including its official website, mail server, and cyber protection services. Furthermore, the hackers managed to access and download over 10 gigabytes of data from M9com's mail server and client databases. To make matters worse, they made this stolen information publicly accessible via the Tor browser. [...]

Based on the nature of the attack on M9com, it appears that when the hackers hit Moscow, they were able to gain access to the back-end operations of the company. This allowed them to effectively delete data from the servers, similar to what occurred in the Kyivstar incident. It is worth noting that this type of attack, which involves directly targeting and compromising the servers, is less common compared to the more frequently observed distributed denial-of-service (DDoS) attacks. DDoS attacks overwhelm a system by inundating it with automated requests, causing the service to become inaccessible.