# This Week's Releases from OpenAI | EP 148

Source: https://www.youtube.com/watch?v=F3OtgJorWx8
Recap page: https://rapidrecap.app/video/F3OtgJorWx8
Generated: 2025-08-08T20:02:52.053+00:00

---
## Quick Overview

OpenAI released GPT-5, which Sam Altman described as a major upgrade and a significant step towards AGI, comparing it to talking to a "PhD in a subject" versus a "high school student" for GPT-3 and a "college student" for GPT-4, though it does not continuously learn. Concurrently, Amazon launched an overhauled Alexa Plus, integrating generative AI, which offers a more conversational and personalized experience but suffers from latency, broken basic functions like alarms, and persistent upselling, leading to a mixed reception and highlighting the challenges of integrating new AI into legacy systems.

**Key Points:**
- OpenAI released GPT-5, which CEO Sam Altman claims is a major upgrade and a significant step towards AGI, stating, "GPT5 he said is uh the first time it feels like talking to an expert, someone you know who has a PhD in a subject."
- Altman clarified that GPT-5 does not continuously learn, which he considers a prerequisite for AGI, and that while previous models felt like talking to a "high school student" (GPT-3) or "college student" (GPT-4), GPT-5 feels like conversing with an expert.
- Amazon launched Alexa Plus, an AI-powered overhaul of its virtual assistant, offering more conversational capabilities and integrations, but early users reported issues like latency, broken alarm functions, and persistent product advertising.
- One host described the Alexa Plus experience as infuriating, stating, "I was just like, "Why did I just spend $90 to have a permanent rotating advertisement for household products like on my desk? That is so weird."
- Despite improvements in voice and conversational flow, Alexa Plus was noted to regress on basic functions, with one host experiencing, "the new Alexa Plus update seems to have broken Alexa's ability to reliably uh set and cancel alarms, which is a core thing that I use this product for."
- The hosts concluded that integrating advanced AI like LLMs into legacy systems is a significant challenge, leading to a "messy fit" and unreliability, with one host stating, "The finest minds in the world who are working on this stuff actually don't know how to do this yet."
- OpenAI also released two open-source models, aiming to compete with other AI providers, and claims GPT-5 has a reduced hallucination rate of around 1%, though hosts remain cautious and plan further testing.

**Context:** The "Hard Fork" podcast episode discusses recent advancements and challenges in the AI landscape, focusing on OpenAI's latest flagship model, GPT-5, and Amazon's significant update to its virtual assistant, Alexa Plus. The conversation highlights the rapid pace of AI development, the ongoing race towards Artificial General Intelligence (AGI), and the practical difficulties of integrating advanced AI technologies into existing consumer products and infrastructure.

## Detailed Analysis

The podcast "Hard Fork" discusses two major AI developments: OpenAI's release of GPT-5 and Amazon's revamped Alexa Plus. Regarding GPT-5, the hosts note that Sam Altman described it as a significant step towards AGI, a major upgrade that makes previous models "quite miserable" by comparison, likening GPT-3 to a "high school student," GPT-4 to a "college student," and GPT-5 to an "expert with a PhD." However, Altman clarified that GPT-5 does not continuously learn, a key characteristic of AGI. OpenAI also released two open-source models, aiming to compete with models from China. The hosts express a need to test GPT-5 thoroughly, with one host planning to use their "Roosebench" tests, and note that the marketing hype for new models often leads to questions about whether they enable truly new capabilities. They also discuss OpenAI's claims of improved reliability, reduced hallucinations (down to around 1%), and "safe completions" where the model offers a safer alternative to a problematic request. The discussion touches on OpenAI's business strategy, including aggressive pricing for developers, and the broader AI landscape, with one host noting that market predictions shifted from OpenAI to Google immediately after the GPT-5 announcement, suggesting it was seen as evolutionary rather than revolutionary. Separately, the podcast reviews Amazon's Alexa Plus, an overhaul of the Alexa virtual assistant powered by generative AI, possibly using Amazon's models and Anthropic's Claude. While it offers a more conversational and fluid voice, allows for multi-turn conversations without constant wake words, and integrates with services like Uber and OpenTable, early access users report significant issues. These include persistent advertising for Amazon products on devices like the Echo Show, broken core functionalities such as setting alarms, increased latency compared to older Alexa versions, occasional hallucinations, and difficulty orchestrating tasks or understanding user intent. One host found the experience infuriating, likening the devices to "permanent rotating advertisement[s]," while the other noted that while some capabilities are cool, basic functions are regressing, making it a "two steps forward, one step back" situation. Both hosts emphasize the difficulty of integrating LLMs into legacy systems, drawing parallels to Apple's Siri challenges, and suggest that the finest minds in AI may not yet have solved how to effectively merge these technologies, leading to a messy fit across the economy.

### OpenAI GPT-5 Release

- Sam Altman calls it a major upgrade and step towards AGI
- Compares GPT-5 to an expert with a PhD, GPT-4 to a college student, GPT-3 to a high school student
- GPT-5 does not continuously learn, a key AGI characteristic
- OpenAI also released two open-source models
- Claims of reduced hallucinations (around 1%) and improved reliability
- Pricing for developers is competitive, matching Google's Gemini 2.5 Pro

### Alexa Plus Overhaul

- Generative AI integration aims for more conversational and personalized experience
- Improved voice fluidity and multi-turn conversations without repeated wake words
- Integrations with services like Uber and OpenTable show promise, with some early successes
- Core functionalities like alarms appear broken, exhibiting latency and unreliability
- Persistent advertising for Amazon products on devices is a significant drawback
- Hallucinations and difficulty understanding complex user intent are reported issues
- Hosts find the integration of LLMs into legacy systems challenging and often unreliable, a trend seen across the economy

### AI Development Challenges

- "Finest minds" may not yet know how to effectively integrate LLMs into legacy systems
- Messy fits occur when jamming new AI into old systems, affecting various economic sectors
- User patience for hardware with unfulfilled promises is low, especially for smart home devices

### Host Personal Experiences

- One host found Alexa Plus infuriating due to constant ads and broken core functions
- Another host experienced latency, hallucinations, and issues with task orchestration
- Both hosts express a need for further testing but report negative first impressions of Alexa Plus's reliability