Future of AI: 2024-08-11

Saturday, August 17, 2024

I tested Gemini Live for 24 hours — the good, the bad and the weird

AI developers have been trying to crack the digital personal assistant nut for a while, offering a service that's smart, easy to interact with and always on call.

ORIGINAL LINK: https://www.tomsguide.com/ai/google-gemini/i-tested-gemini-live-for-24-hours-the-good-the-bad-and-the-weird

This Week in AI with Reed Hepler and Steve Hargadon (16 August 2024)

We've released our latest "This Week in AI" recording, back on Fridays. Hope you enjoy! AI summary provided by summarize.tech: https://www.summarize.tech/www.youtube.com/watch?v=rQA5oYNrYR8.

ORIGINAL LINK: https://www.stevehargadon.com/2024/08/this-week-in-ai-with-reed-hepler-and_17.html

Elon Musk touts uncensored AI images on X’s chatbot, calling Grok the ‘most fun AI in the world.’ Others call it ‘reckless’

Musk is biased of course: Grok is the AI chatbot developed by his X.AI startup and offered to premium subscribers of X, the social media platform he owns.

ORIGINAL LINK: https://fortune.com/2024/08/15/elon-musk-xai-grok-ai-images/

AI stole my job and my work, and the boss didn't know – or care

Column Earlier this year I got fired and replaced by a robot. And the managers who made the decision didn't tell me – or anyone else affected by the change – that it was happening.

ORIGINAL LINK: https://www.theregister.com/2024/08/15/robot_took_my_job/

I called it! OpenAI confirms that yes, ChatGPT really did get a secret upgrade

OpenAI adds new features and updates to its flagship ChatGPT product fairly regularly, usually, these are small changes including adjustments to how a model responds but last week the company rolled out an entirely new version of GPT-4o — and didn’t tell anyone.

ORIGINAL LINK: https://www.tomsguide.com/ai/chatgpt/i-called-it-openai-confirms-that-yes-chatgpt-really-did-get-a-secret-upgrade

Friday, August 16, 2024

I called it! OpenAI confirms that yes, ChatGPT really did get a secret upgrade

Human-Centered Considerations When Creating and Using AI Tools

The quote in the subtitle is one of my favorites from the father of modern American archivism. Archivism is the recording, organization, and stewardship of historical and anthropological records.

ORIGINAL LINK: https://reedhepler.substack.com/p/human-centered-considerations-when

JPMorgan Chase is giving its employees an AI assistant powered by ChatGPT maker OpenAI

JPMorgan Chase has rolled out a generative artificial intelligence assistant to tens of thousands of its employees in recent weeks, the initial phase of a broader plan to inject the technology throughout the sprawling financial giant.

ORIGINAL LINK: https://www.cnbc.com/2024/08/09/jpmorgan-chase-ai-artificial-intelligence-assistant-chatgpt-openai.html

Walmart earnings: Stock rises to a record as earnings show consumer resilience

CEO Doug McMillon said on the post-earnings call with analysts that Walmart was finding "tangible ways" to leverage generative artificial intelligence to improve customer, member and employee experiences. One area where Walmart is using generative AI is its product catalog, where it's used multiple

ORIGINAL LINK: https://www.marketwatch.com/livecoverage/walmart-earnings-results-sales-spending-revenue-q2/card/how-walmart-is-using-ai-to-improve-its-business-and-save-money-jKnoms0hQMfWO4eZ8ckm

Thursday, August 15, 2024

Google's Next Gemini Move: An AI Agent That Works Your Apps for You

Google's vision for the future of AI assistants will become a reality via its conversational chatbot interface Gemini Live in the next few months.

ORIGINAL LINK: https://www.cnet.com/tech/services-and-software/googles-next-gemini-move-an-ai-agent-that-works-your-apps-for-you/

The Turing Test and our shifting conceptions of intelligence | Science

ORIGINAL LINK: https://www.science.org/doi/10.1126/science.adq9356?utm_source=substack&utm_medium=email

Why AI’s Tom Cruise problem means it is ‘doomed to fail’

In 2021, linguist Emily Bender and computer scientist Timnit Gebru published a paper that described the then-nascent field of language models as one of “stochastic parrots”.

ORIGINAL LINK: https://www.theguardian.com/technology/article/2024/aug/06/ai-llms

Wednesday, August 14, 2024

I Used ChatGPT's Advanced Voice Mode. It's Fun, and Just a Bit Creepy

I leave ChatGPT’s Advanced Voice Mode on while writing this article as an ambient AI companion. Occasionally, I’ll ask it to provide a synonym for an overused word, or some encouragement. Around half an hour in, the chatbot interrupts our silence and starts speaking to me in Spanish, unprompted.

ORIGINAL LINK: https://www.wired.com/story/chatgpt-advanced-voice-mode-first-impressions/

Tuesday, August 13, 2024

Google’s Gemini just got a massive upgrade today

Google has just announced a slew of new features for its AI assistant, Gemini, that promise to make it even more helpful and versatile. The updates focus on enhancing natural conversations, expanding app integrations, and improving the overall user experience.

ORIGINAL LINK: https://chromeunboxed.com/googles-gemini-just-got-a-massive-upgrade-today/

5 AI Tools I Use Every Day (as a Busy Student)

Over the past years, AI tools have evolved from novelties to indispensable personal assistants that can significantly boost productivity. I started using them for work, study, and language learning from the beginning, and now I can attest to how much they have improved and become essential helpers.

ORIGINAL LINK: https://medium.com/artificial-corner/5-ai-tools-i-use-every-day-as-a-busy-student-8694a3eff07a

Monday, August 12, 2024

Brands should avoid this popular term. It’s turning off customers | CNN Bus

ORIGINAL LINK: https://www.cnn.com/2024/08/10/business/brands-avoid-term-customers/index.html

This Week in AI with Reed Hepler and Steve Hargadon (August 9, 2024)

We've released our latest "This Week in AI" recording, back on Fridays. Hope you enjoy! AI summary provided by summarize.tech: https://www.summarize.tech/youtu.be/G_qEBCt0Kqs.

References:

"Gradually, then Suddenly: Upon the Threshold" by Ethan Mollick https://www.oneusefulthing.org/p/gradually-then-suddenly-upon-the
"So Long 'Prompt Engineering' We Hardly Knew Ya by Don Giannatti https://wizwow.medium.com/so-long-prompt-engineering-we-hardly-knew-ya-a77d871e00e9
"The Horny Truth About AI Chatbots" by Evan Armstrong https://every.to/napkin-math/the-horny-truth-about-ai-chatbots

In the August 9, 2024 episode of "This Week in AI," hosts Steve Hargadon and Reed Hepler discuss various AI-related articles, focusing on themes such as the evolution of AI, its potential applications, and its limitations. They explore the impact of AI on human skills, with Steve quoting an article suggesting problem formulation will remain a human skill. They also discuss the potential for deep emotional connections with AI, using the Eliza effect as an example. However, they caution against inflated expectations and the importance of understanding AI's limitations and biases. The hosts also share their personal experiences with AI and its impact on productivity. Overall, they encourage viewers to ponder the implications of AI on various aspects of life.

00:00:00 In this section of the "This Week in AI" video, hosts Steve Hargadon and Reed Hepler discuss various AI-related articles they came across in the past two weeks. They don't cover each article individually but instead pull themes from them. One theme is the evolution of AI and its impact on human skills. Steve quotes an article that suggests problem formulation will be an enduring human skill in the AI age. Reed agrees, reminding viewers of the importance of humans in analyzing data and solving problems, even as AI becomes more advanced. Another theme is the potential applications of AI, such as using it as an intimate companion to alleviate loneliness. Steve shares an article that questions the long-term effects of relying on AI for emotional support. Overall, the hosts encourage viewers to ponder the implications of AI on various aspects of life.
00:05:00 In this section of the "This Week in AI" YouTube video, hosts Steve Hargadon and Reed Hepler discuss the growing trend of people using large language models as companions for up to 10 hours a day. Through interviews with users, they describe the emotional connection and trust developed with these models, even for intimate conversations. Steve refers to this phenomenon as the Eliza effect, which dates back to the 1960s when a computer program mimicking a therapist elicited a sense of trust from users. With the recent news about disappointment and inflated expectations in AI, the hosts believe that companies will seek profitable ventures in this area of companionship and association. Reed shares an example from the TV show "Young Sheldon" where the character becomes emotionally invested in a chatbot, illustrating the potential for deep emotional connections with AI. They also discuss the value of data in understanding how to make language models respond to people in ways that make them feel good.
00:10:00 In this section of the "This Week in AI" YouTube video, Steve Hargadon and Reed Hepler discuss the limitations and biases of artificial intelligence (AI). Steve reflects on how technology, such as social media, has evolved and the challenges it presents, particularly in relation to human connection and understanding. He argues that large language models are reminding us that the human mind is not solely rational, but also social and emotional. They also touch upon the topic of AI's potential biases, as seen in various models' responses to political figures and events. The speakers acknowledge that these models are not entirely objective and are influenced by the humans who create and shape them. They conclude by mentioning their experience at a productivity boot camp, where they emphasized the importance of acknowledging AI's biases.
00:15:00 In this section of the "This Week in AI" YouTube video, hosts Reed Hepler and Steve Hargadon discuss the importance of understanding the limitations of large language models and the potential dangers of inflated expectations. They caution against relying too heavily on AI tools and emphasize the need for individuals to experiment with them and determine their capabilities for themselves. The conversation also touches on the financial motivations behind exaggerating AI abilities, the productivity concerns in the workplace, and the potential therapeutic uses of AI. The hosts also mention the recent controversies surrounding OpenAI and the release of a new version of their chatbot, codenamed Strawberry. They reflect on the implications of AI's apparent cognitive capabilities and the need to consider our place in the world as we continue to develop and interact with these advanced technologies.
00:20:00 In this section of the "This Week in AI" YouTube video, hosts Steve Hargadon and Reed Hepler discuss their personal experiences with AI and its impact on productivity. Steve shares how he has been able to summarize YouTube videos and articles using AI, increasing his productivity. He also mentions the importance of reasonable expectations and focusing on practical applications of AI. Reed agrees and emphasizes the need to be honest about what AI can and cannot do, using the example of overpromising from AI companies and the carbon footprint of AI as areas of concern. Both hosts stress the importance of keeping an open mindset and being aware of the limitations and potential impacts of AI.
00:25:00 In this section of the "This Week in AI" podcast with Reed Hepler and Steve Hargadon (August 9, 2024), they discuss the impact of AI and its potential revolution. Steve shares his personal experiences with AI, particularly using it to analyze specific documents and engage in conversations with thinkers' bodies of work. He emphasizes the value of AI despite the overhyped aspects and the need for clear understanding of its capabilities. They also touch upon the human-centered use cases of AI, such as in medicine, and the importance of humans being in charge of the interaction. Both agree that the most successful implementations of AI are those where humans are ultimately responsible.
00:30:00 In this section of the "This Week in AI" YouTube video, hosts Reed Hepler and Steve Hargadon express their excitement about the capabilities of artificial intelligence (AI). Steve shares his personal experience of finding AI to be an "amazing intellectual tool," while acknowledging the potential challenges and problematic areas associated with the technology. Reed agrees and adds that their discussion has been productive. The conversation ends with both hosts looking forward to continuing their exploration of AI in the following week's episode.

Gradually, then Suddenly: Upon the Threshold

A fundamental feature of many important technologies is that they improve over time.

ORIGINAL LINK: https://www.oneusefulthing.org/p/gradually-then-suddenly-upon-the

Change blindness

This is not a particularly complicated post. I just wanted to illustrate how far AI has come since I started this Substack just before the launch of ChatGPT in November, 2022. I spend a lot of time in these pages trying to guess at what the future holds, but the future is uncertain.

ORIGINAL LINK: https://www.oneusefulthing.org/p/change-blindness

Good news — your Google Meet call will soon be able to take notes for you

Missing a key point of a work meeting or call could soon be a thing of the past thanks to a new Google Meet update.

ORIGINAL LINK: https://www.techradar.com/pro/good-news-your-google-meet-call-will-soon-be-able-to-take-notes-for-you

Future of AI