AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp

AI is getting so capable that unless you're doing some seriously complex tasks, new models are starting to have diminishing returns. For years AI companies were primarily worried about intelligence, but we think that it's time they start to improve in other areas. AI has a lot of room to improve in token efficiency, UX, and other consumer-friendly practices that would help them appeal to the less tech-savvy people out there. In this edition of the Web News, Matt and Mike discussed their opinion the state of AI beyond intelligence. What do you think AI should improve upon beyond simply making them smarter with each model? #techtok #ai #chatgpt #softwareengineer #fyp

Is the AI bubble getting ready to pop? Recently, OpenAI adjusted their ChatGPT subscriptions such that some users will be getting less usage for the same price. Just this week, Matt had 3TB of OneDrive space slashed from his M365 Family account, yet the price remains the same as before. As components (RAM, SSDs, etc.) get more expensive, it's only natural that we'll see SaaS models give out less features for the same or higher prices. The thing is, most of the increased cost is due to an immense amount of AI spend - Anthropic, OpenAI, and others are dumping money into increasing levels of compute so that they can remain competitive in an ever-changing landscape. All the while, the AI market has been subsidized by investors looking to get rich off of what they speculate will be a massive increase in value one day. But it's been years and AI still has yet to reach some of its major milestones, including AGI. If investors start to get cold feet, how long will this subsidized pricing continue? And, as other segments of the market continue to face higher expenses due to component price inflation - they'll naturally pass those expenses onto consumers. #ai #chatgpt #softwareengineer #techtok #fyp

Some technical interviews are incorporating real-world or practical testing periods where you're given a task with free reign on how to solve it. Mike believes that if the job posting or interviewer has mentioned AI or agentic coding at all, you should use AI to help you solve any roadblocks you might hit. You might feel obligated to use more traditional troubleshooting like Googling, Stack Overflow, or Reddit posts - but they might want to see how you would use AI in the actual job. #technicalinterview #techjobs #softwareengineer #ai #fyp