×
OpenAI gets caught vibe graphing

OpenAI gets caught vibe graphing

Tech-news , ,

During its big GPT-5 livestream on Thursday, OpenAI showed off a few charts that made the model seem quite impressive — but if you look closely, some graphs were a little bit off.

In one, ironically showing how well GPT-5 does in “deception evals across models,” the scale is all over the place. For “coding deception,” for example, the chart shown onstage says GPT-5 with thinking apparently gets a 50.0 percent deception rate, but that’s compared to OpenAI’s smaller 47.4 percent o3 score which somehow has a larger bar. OpenAI appears to have accurate numbers for this chart in its GPT-5 blog post, however, where GPT-5’s deception rate is labeled as 16.5 percent.

With this chart, OpenAI showed onstage that one of GPT-5’s scores is lower than o3’s but is shown with a bigger bar. In this same chart, o3 and GPT-4o’s scores are different but shown with equally-sized bars. It was bad enough that CEO Sam Altman commented on it, calling it a “mega chart screwup,” though he noted that a correct version is in OpenAI’s blog post.

An OpenAI marketing staffer also apologized, saying, “We fixed the chart in the blog guys, apologies for the unintentional chart crime.”

OpenAI didn’t immediately respond to a request for comment. And while it’s unclear if OpenAI used GPT-5 to actually make the charts, it’s still not a great look for the company on its big launch day — especially when it is touting the “significant advances in reducing hallucinations” with its new model.

Source link
#OpenAI #caught #vibe #graphing

Tag

Related Posts

Focus and productivity apps abound, all to help you stave off the many distractions coming from your phone. Or the annoying people at your open-office desk. Digital well-being tools can silence notifications, limit apps like TikTok and Instagram, and help you focus on the task at hand. But you can also turn them off very easily as soon as you feel like you haven’t endlessly scrolled enough.

This is where Flipper’s Busy Bar comes in, a hardware clock with an LED screen that doubles as a clock and a dedicated timer. Slap the big button in the middle, and the screen displays a bright red “BUSY” sign or another message that lets the people around you know you’re, well, busy. (Maybe try “GO AWAY” or “GET OUT OF MY ROOM, MOM.”) The bar goes on sale today and costs $249.

“How do you let people know politely, yet firmly, that you don’t want to be disturbed?” says Callum Tennent, a creative writer at Flipper. “We decided the politest way to do it was a massive red light on your desk.”

Image may contain Computer Hardware Electronics Hardware Monitor Screen Mobile Phone and Phone

Courtesy of Flipper Devices

Flipper Devices made the Flipper Zero, a $200 portable hacking tool that got big on TikTok in 2022 for using a Tamagotchi-esque dolphin character to detect wireless frequencies and potentially break RFID-controlled locks. It was a device that raised a variety of security concerns. Canada proposed a ban on the device out of fear that it might enable car thefts. In 2023, the US Customs and Border Protection seized 15,000 Flipper Zero devices, then ultimately released them. Flipper is currently working on another model, the Flipper One, that has even more advanced capabilities.

In between those more controversial devices comes the Busy Bar. The bar also works with the separate Busy app, yet another one of those productivity and focus tools living on your phone. What it doesn’t have is the capability of hacking anything. “It’s being made by us here at Flipper, but there’s no real connection to them,” Tennent says. “They’re totally disconnected products.”

Fundamentally, the Busy Bar is a pricey “On Air” light. It offers many of the same productivity capabilities that are likely already baked into your phone’s operating system—like blocking notifications on your phone. But Flipper is making the case that—much like the Brick, a hardware gadget you tap to block access to certain apps—having a hardware option to shut off the distractions around you is meaningfully different than just trying to use software productivity tools on your device.

#Busy #Bar #Gadget #People #Leavehacks,gadgets,smart home,apps,distractions,productivity,design">The Busy Bar Is a Gadget to Get People to Leave You AloneFocus and productivity apps abound, all to help you stave off the many distractions coming from your phone. Or the annoying people at your open-office desk. Digital well-being tools can silence notifications, limit apps like TikTok and Instagram, and help you focus on the task at hand. But you can also turn them off very easily as soon as you feel like you haven’t endlessly scrolled enough.This is where Flipper’s Busy Bar comes in, a hardware clock with an LED screen that doubles as a clock and a dedicated timer. Slap the big button in the middle, and the screen displays a bright red “BUSY” sign or another message that lets the people around you know you’re, well, busy. (Maybe try “GO AWAY” or “GET OUT OF MY ROOM, MOM.”) The bar goes on sale today and costs 9.“How do you let people know politely, yet firmly, that you don’t want to be disturbed?” says Callum Tennent, a creative writer at Flipper. “We decided the politest way to do it was a massive red light on your desk.”Courtesy of Flipper DevicesFlipper Devices made the Flipper Zero, a 0 portable hacking tool that got big on TikTok in 2022 for using a Tamagotchi-esque dolphin character to detect wireless frequencies and potentially break RFID-controlled locks. It was a device that raised a variety of security concerns. Canada proposed a ban on the device out of fear that it might enable car thefts. In 2023, the US Customs and Border Protection seized 15,000 Flipper Zero devices, then ultimately released them. Flipper is currently working on another model, the Flipper One, that has even more advanced capabilities.In between those more controversial devices comes the Busy Bar. The bar also works with the separate Busy app, yet another one of those productivity and focus tools living on your phone. What it doesn’t have is the capability of hacking anything. “It’s being made by us here at Flipper, but there’s no real connection to them,” Tennent says. “They’re totally disconnected products.”Fundamentally, the Busy Bar is a pricey “On Air” light. It offers many of the same productivity capabilities that are likely already baked into your phone’s operating system—like blocking notifications on your phone. But Flipper is making the case that—much like the Brick, a hardware gadget you tap to block access to certain apps—having a hardware option to shut off the distractions around you is meaningfully different than just trying to use software productivity tools on your device.#Busy #Bar #Gadget #People #Leavehacks,gadgets,smart home,apps,distractions,productivity,design

apps abound, all to help you stave off the many distractions coming from your phone. Or the annoying people at your open-office desk. Digital well-being tools can silence notifications, limit apps like TikTok and Instagram, and help you focus on the task at hand. But you can also turn them off very easily as soon as you feel like you haven’t endlessly scrolled enough.

This is where Flipper’s Busy Bar comes in, a hardware clock with an LED screen that doubles as a clock and a dedicated timer. Slap the big button in the middle, and the screen displays a bright red “BUSY” sign or another message that lets the people around you know you’re, well, busy. (Maybe try “GO AWAY” or “GET OUT OF MY ROOM, MOM.”) The bar goes on sale today and costs $249.

“How do you let people know politely, yet firmly, that you don’t want to be disturbed?” says Callum Tennent, a creative writer at Flipper. “We decided the politest way to do it was a massive red light on your desk.”

Image may contain Computer Hardware Electronics Hardware Monitor Screen Mobile Phone and Phone

Courtesy of Flipper Devices

Flipper Devices made the Flipper Zero, a $200 portable hacking tool that got big on TikTok in 2022 for using a Tamagotchi-esque dolphin character to detect wireless frequencies and potentially break RFID-controlled locks. It was a device that raised a variety of security concerns. Canada proposed a ban on the device out of fear that it might enable car thefts. In 2023, the US Customs and Border Protection seized 15,000 Flipper Zero devices, then ultimately released them. Flipper is currently working on another model, the Flipper One, that has even more advanced capabilities.

In between those more controversial devices comes the Busy Bar. The bar also works with the separate Busy app, yet another one of those productivity and focus tools living on your phone. What it doesn’t have is the capability of hacking anything. “It’s being made by us here at Flipper, but there’s no real connection to them,” Tennent says. “They’re totally disconnected products.”

Fundamentally, the Busy Bar is a pricey “On Air” light. It offers many of the same productivity capabilities that are likely already baked into your phone’s operating system—like blocking notifications on your phone. But Flipper is making the case that—much like the Brick, a hardware gadget you tap to block access to certain apps—having a hardware option to shut off the distractions around you is meaningfully different than just trying to use software productivity tools on your device.

#Busy #Bar #Gadget #People #Leavehacks,gadgets,smart home,apps,distractions,productivity,design">The Busy Bar Is a Gadget to Get People to Leave You Alone

Focus and productivity apps abound, all to help you stave off the many distractions coming from your phone. Or the annoying people at your open-office desk. Digital well-being tools can silence notifications, limit apps like TikTok and Instagram, and help you focus on the task at hand. But you can also turn them off very easily as soon as you feel like you haven’t endlessly scrolled enough.

This is where Flipper’s Busy Bar comes in, a hardware clock with an LED screen that doubles as a clock and a dedicated timer. Slap the big button in the middle, and the screen displays a bright red “BUSY” sign or another message that lets the people around you know you’re, well, busy. (Maybe try “GO AWAY” or “GET OUT OF MY ROOM, MOM.”) The bar goes on sale today and costs $249.

“How do you let people know politely, yet firmly, that you don’t want to be disturbed?” says Callum Tennent, a creative writer at Flipper. “We decided the politest way to do it was a massive red light on your desk.”

Image may contain Computer Hardware Electronics Hardware Monitor Screen Mobile Phone and Phone

Courtesy of Flipper Devices

Flipper Devices made the Flipper Zero, a $200 portable hacking tool that got big on TikTok in 2022 for using a Tamagotchi-esque dolphin character to detect wireless frequencies and potentially break RFID-controlled locks. It was a device that raised a variety of security concerns. Canada proposed a ban on the device out of fear that it might enable car thefts. In 2023, the US Customs and Border Protection seized 15,000 Flipper Zero devices, then ultimately released them. Flipper is currently working on another model, the Flipper One, that has even more advanced capabilities.

In between those more controversial devices comes the Busy Bar. The bar also works with the separate Busy app, yet another one of those productivity and focus tools living on your phone. What it doesn’t have is the capability of hacking anything. “It’s being made by us here at Flipper, but there’s no real connection to them,” Tennent says. “They’re totally disconnected products.”

Fundamentally, the Busy Bar is a pricey “On Air” light. It offers many of the same productivity capabilities that are likely already baked into your phone’s operating system—like blocking notifications on your phone. But Flipper is making the case that—much like the Brick, a hardware gadget you tap to block access to certain apps—having a hardware option to shut off the distractions around you is meaningfully different than just trying to use software productivity tools on your device.

#Busy #Bar #Gadget #People #Leavehacks,gadgets,smart home,apps,distractions,productivity,design

SAVE $20: As of June 29, the Google TV Streamer 4K is on sale for $79.99 at Amazon. That’s a 20% discount on the list price.

$79.99 at Amazon
$99.99 Save $20

 

Looking for a simple way to upgrade your streaming setup? Sounds like it’s time to invest in a streaming stick.

These handy pieces of tech plug straight into your TV to give you easy and quick access to apps, streaming platforms, and live TV options. And as of June 29, the Google TV Streamer 4K is on sale for $20 off the list price, now down to $79.99.

Mashable Deals

By signing up, you agree to receive recurring automated SMS marketing messages from Mashable Deals at the number provided. Msg and data rates may apply. Up to 2 messages/day. Reply STOP to opt out, HELP for help. Consent is not a condition of purchase. See our Privacy Policy and Terms of Use.

This streaming device also gives you great quality. It supports up to 4K HDR with Dolby Vision and works with Dolby Atmos-compatible speakers for more immersive, 3D-style sound. Thanks to its powerful processor, it runs fast, and it has double the memory compared to the previous generation of this device. It also includes a redesigned remote with voice control and smart home control. You can even find the remote if lost by making it ring.

And of course, you’ll have access to all the major streaming apps like Netflix, Prime Video, and more. Plus, over 800 free live TV channels from apps like Pluto TV and Tubi are at your fingertips.

This Google Streamer deal is available at Amazon right now.

#streaming #stick #deal #Save #Google #Streamer">Best streaming stick deal: Save on Google TV Streamer 4K SAVE : As of June 29, the Google TV Streamer 4K is on sale for .99 at Amazon. That’s a 20% discount on the list price. .99 at Amazon .99 Save   Looking for a simple way to upgrade your streaming setup? Sounds like it’s time to invest in a streaming stick. These handy pieces of tech plug straight into your TV to give you easy and quick access to apps, streaming platforms, and live TV options. And as of June 29, the Google TV Streamer 4K is on sale for off the list price, now down to .99. Mashable Deals By signing up, you agree to receive recurring automated SMS marketing messages from Mashable Deals at the number provided. Msg and data rates may apply. Up to 2 messages/day. Reply STOP to opt out, HELP for help. Consent is not a condition of purchase. See our Privacy Policy and Terms of Use. This streaming device also gives you great quality. It supports up to 4K HDR with Dolby Vision and works with Dolby Atmos-compatible speakers for more immersive, 3D-style sound. Thanks to its powerful processor, it runs fast, and it has double the memory compared to the previous generation of this device. It also includes a redesigned remote with voice control and smart home control. You can even find the remote if lost by making it ring.And of course, you’ll have access to all the major streaming apps like Netflix, Prime Video, and more. Plus, over 800 free live TV channels from apps like Pluto TV and Tubi are at your fingertips. Mashable Deals By signing up, you agree to receive recurring automated SMS marketing messages from Mashable Deals at the number provided. Msg and data rates may apply. Up to 2 messages/day. Reply STOP to opt out, HELP for help. Consent is not a condition of purchase. See our Privacy Policy and Terms of Use. This Google Streamer deal is available at Amazon right now. #streaming #stick #deal #Save #Google #Streamer

Google TV Streamer 4K is on sale for $79.99 at Amazon. That’s a 20% discount on the list price.

$79.99 at Amazon
$99.99 Save $20

 

Looking for a simple way to upgrade your streaming setup? Sounds like it’s time to invest in a streaming stick.

These handy pieces of tech plug straight into your TV to give you easy and quick access to apps, streaming platforms, and live TV options. And as of June 29, the Google TV Streamer 4K is on sale for $20 off the list price, now down to $79.99.

Mashable Deals

By signing up, you agree to receive recurring automated SMS marketing messages from Mashable Deals at the number provided. Msg and data rates may apply. Up to 2 messages/day. Reply STOP to opt out, HELP for help. Consent is not a condition of purchase. See our Privacy Policy and Terms of Use.

This streaming device also gives you great quality. It supports up to 4K HDR with Dolby Vision and works with Dolby Atmos-compatible speakers for more immersive, 3D-style sound. Thanks to its powerful processor, it runs fast, and it has double the memory compared to the previous generation of this device. It also includes a redesigned remote with voice control and smart home control. You can even find the remote if lost by making it ring.

And of course, you’ll have access to all the major streaming apps like Netflix, Prime Video, and more. Plus, over 800 free live TV channels from apps like Pluto TV and Tubi are at your fingertips.

This Google Streamer deal is available at Amazon right now.

#streaming #stick #deal #Save #Google #Streamer">Best streaming stick deal: Save $20 on Google TV Streamer 4K

SAVE $20: As of June 29, the Google TV Streamer 4K is on sale for $79.99 at Amazon. That’s a 20% discount on the list price.

$79.99 at Amazon
$99.99 Save $20

 

Looking for a simple way to upgrade your streaming setup? Sounds like it’s time to invest in a streaming stick.

These handy pieces of tech plug straight into your TV to give you easy and quick access to apps, streaming platforms, and live TV options. And as of June 29, the Google TV Streamer 4K is on sale for $20 off the list price, now down to $79.99.

Mashable Deals

By signing up, you agree to receive recurring automated SMS marketing messages from Mashable Deals at the number provided. Msg and data rates may apply. Up to 2 messages/day. Reply STOP to opt out, HELP for help. Consent is not a condition of purchase. See our Privacy Policy and Terms of Use.

This streaming device also gives you great quality. It supports up to 4K HDR with Dolby Vision and works with Dolby Atmos-compatible speakers for more immersive, 3D-style sound. Thanks to its powerful processor, it runs fast, and it has double the memory compared to the previous generation of this device. It also includes a redesigned remote with voice control and smart home control. You can even find the remote if lost by making it ring.

And of course, you’ll have access to all the major streaming apps like Netflix, Prime Video, and more. Plus, over 800 free live TV channels from apps like Pluto TV and Tubi are at your fingertips.

This Google Streamer deal is available at Amazon right now.

#streaming #stick #deal #Save #Google #Streamer
China’s Zhipu AI (Z.ai) released its open-weight GLM-5.2, and some researchers have claimed that it matches Mythos in certain bug-finding and cybersecurity scenarios. While GLM lags behind models from Anthropic and OpenAI in other, more general tasks, it seems that China has dramatically reduced the gap in the capabilities between its models and those of the US.

This level of advancement is particularly concerning to the US government, which has worked to restrict China’s access to powerful models like Anthropic’s Mythos and Fable, as well as the hardware necessary to train and run them. The Trump administration views Mythos and other advanced AI models capable of identifying vulnerabilities as serious national security threats. Recently, OpenAI unveiled GPT-5.6, which has also raised concerns about its potential for misuse and has limited access to it.

Because GLM is an open-weight model, it can be downloaded and run by anyone on readily available hardware. That gives it great flexibility and allows power users deep access, but it also makes it ripe for abuse by bad actors who can run it with little oversight.

#Chinas #Z.ai #claims #match #Mythos #cybersecurityAI,News,Policy,Politics,Security,Tech">China’s Z.ai claims it can match Mythos on cybersecurityChina’s Zhipu AI (Z.ai) released its open-weight GLM-5.2, and some researchers have claimed that it matches Mythos in certain bug-finding and cybersecurity scenarios. While GLM lags behind models from Anthropic and OpenAI in other, more general tasks, it seems that China has dramatically reduced the gap in the capabilities between its models and those of the US.This level of advancement is particularly concerning to the US government, which has worked to restrict China’s access to powerful models like Anthropic’s Mythos and Fable, as well as the hardware necessary to train and run them. The Trump administration views Mythos and other advanced AI models capable of identifying vulnerabilities as serious national security threats. Recently, OpenAI unveiled GPT-5.6, which has also raised concerns about its potential for misuse and has limited access to it.Because GLM is an open-weight model, it can be downloaded and run by anyone on readily available hardware. That gives it great flexibility and allows power users deep access, but it also makes it ripe for abuse by bad actors who can run it with little oversight.#Chinas #Z.ai #claims #match #Mythos #cybersecurityAI,News,Policy,Politics,Security,Tech

Z.ai) released its open-weight GLM-5.2, and some researchers have claimed that it matches Mythos in certain bug-finding and cybersecurity scenarios. While GLM lags behind models from Anthropic and OpenAI in other, more general tasks, it seems that China has dramatically reduced the gap in the capabilities between its models and those of the US.

This level of advancement is particularly concerning to the US government, which has worked to restrict China’s access to powerful models like Anthropic’s Mythos and Fable, as well as the hardware necessary to train and run them. The Trump administration views Mythos and other advanced AI models capable of identifying vulnerabilities as serious national security threats. Recently, OpenAI unveiled GPT-5.6, which has also raised concerns about its potential for misuse and has limited access to it.

Because GLM is an open-weight model, it can be downloaded and run by anyone on readily available hardware. That gives it great flexibility and allows power users deep access, but it also makes it ripe for abuse by bad actors who can run it with little oversight.

#Chinas #Z.ai #claims #match #Mythos #cybersecurityAI,News,Policy,Politics,Security,Tech">China’s Z.ai claims it can match Mythos on cybersecurity

China’s Zhipu AI (Z.ai) released its open-weight GLM-5.2, and some researchers have claimed that it matches Mythos in certain bug-finding and cybersecurity scenarios. While GLM lags behind models from Anthropic and OpenAI in other, more general tasks, it seems that China has dramatically reduced the gap in the capabilities between its models and those of the US.

This level of advancement is particularly concerning to the US government, which has worked to restrict China’s access to powerful models like Anthropic’s Mythos and Fable, as well as the hardware necessary to train and run them. The Trump administration views Mythos and other advanced AI models capable of identifying vulnerabilities as serious national security threats. Recently, OpenAI unveiled GPT-5.6, which has also raised concerns about its potential for misuse and has limited access to it.

Because GLM is an open-weight model, it can be downloaded and run by anyone on readily available hardware. That gives it great flexibility and allows power users deep access, but it also makes it ripe for abuse by bad actors who can run it with little oversight.

#Chinas #Z.ai #claims #match #Mythos #cybersecurityAI,News,Policy,Politics,Security,Tech

Post Comment