{"id":93094,"topic":"ai","source":"WRAL","title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","url_hash":"063f09cda74c680c714aa70566bc457f44d1f4f1","author":"","summary":"In Depth: What are countries and companies doing to keep AI from spinning out of control? WRAL","content":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed.\nThat friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.\nOther WRAL Top Stories\nRELATED: Is it too late for North Carolina lawmakers to regulate AI?\nNvidia CEO Jensen Huang said this week that the warnings from artificial intelligence leaders and researchers that AI could pose an existential threat to humanity are overblown.\n“If they believe their company is out of control, then get the company under control,” said Huang in an interview with CNN.\nHuang was referring to the recent X post by a former Anthropic researcher, who wrote “the people building AI earnestly believe that it could kill us all by the end of the decade.”\nDario Amodei, Anthropic’s CEO, recently wrote that AI models could pose \"a catastrophic or existential risk to humanity.\"\nWhat are companies doing to manage the power of AI?\nOn Monday, Nvidia announced a software platform called “Open Shell” that lets companies monitor actions taken by AI agents and enforce policies.\nOpenAI recently announced something similar. The company introduced a technology that tracks and reports when the bots behave badly. This framework for reporting model misalignment comes after an OpenAI model escaped a controlled test on itself and hacked into another AI company called Hugging Face.\nAnthropic has likewise committed to “embedded evaluators.”\nBut this is all self-policing.\nPresident Donald Trump hosted some of the most powerful leaders in the tech industry at the White House on Tuesday, including Huang, Amodei, Mark Zuckerberg and Elon Musk. After the meeting, Trump put an emphasis on self-regulation by the tech companies.\n\"There's a belief that there should be tremendous self-regulation, and we automatically have regulation with the Department of Justice, FBI, all of that. But the self-regulation is very important,\" said Trump.\nWhat about governmental governance?\nTrump recently announced he’s going to put together an AI task force and appoint an AI czar, but no concrete steps have yet been announced.\nIn the meantime, there is talk of early conversations happening between the US and China on establishing some kind of notification mechanism that would alert countries about incidents that pose a threat to national safety.\nIn Europe, countries and companies must abide by the AI Act. Companies that use AI will have to follow varying degrees of rules based on how risky the AI application is.\nBut Sandra Wachter, an Oxford University professor, points out that the EU's AI Act primarily focuses on punishment for individual cases, without really tackling the bigger societal risks that people are increasingly concerned about.\nWhat about global governance?\nThe United Nations has started trying to get global coordination going, but it doesn’t appear we’re close to something like the way they monitor nuclear programs.\nThe International Atomic Energy Agency monitors countries to ensure they are not using nuclear energy to create weapons.\nNot everyone believes there needs to be an IAEA-level of governance for AI.\n“I think the comparison may be unfair. AI, unlike other areas, is in the palms of everyone. Everyone’s using AI. And that’s why I think a multi-layer approach to governance and oversight makes sense for AI,” said Saif Khairat, the Chief AI Officer and Vice Provost for AI at the University of North Carolina at Chapel Hill.\nIt’s true that unlike nuclear technology, AI is everywhere and we can all use it in some way.\nBut like with nuclear technology, America is in an arms race with an adversary about AI: China. And like with nuclear technology, doomsday scenarios about AI are being raised by the very people who are building the technology.\nGreater governance and oversight was a topic of conversation at the United Nations General Assembly last week in New York.\n“History teaches us that humanity’s greatest transformations are never defined by technological breakthroughs alone. Every fundamental leap in capability has demanded an equal leap of governance. We must not allow the cost of today’s hesitation to become tomorrow’s catastrophe,” said Lewis Garseedah Brown II, the representative of Liberia.","image_url":"https://images.wral.com/96296ba0-127c-59a6-93d9-82557c2c0b0b?w=1200&h=630&height=630","lang":"en","published_at":"2026-09-29T23:04:34+00:00","fetched_at":"2026-09-29T23:15:04+00:00","status":"read","starred":0,"extract_state":"ok","summary_auto":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.","cluster_id":null,"extract_retries":0,"extract_error":null,"contract_version":"news_item.v1","format_contract_version":"news_item_formats.v1","dedup_url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 4319 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":4319,"summary_length":267,"usable_text_length":4319,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":4319,"summary_length":267}},"news_item":{"id":93094,"canonical_url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","source_url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","source_name":"WRAL","author":null,"published_at":"2026-09-29T23:04:34+00:00","locale":"en","topic":"ai","tags":[],"rss_summary":"In Depth: What are countries and companies doing to keep AI from spinning out of control? WRAL","full_text":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed.\nThat friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.\nOther WRAL Top Stories\nRELATED: Is it too late for North Carolina lawmakers to regulate AI?\nNvidia CEO Jensen Huang said this week that the warnings from artificial intelligence leaders and researchers that AI could pose an existential threat to humanity are overblown.\n“If they believe their company is out of control, then get the company under control,” said Huang in an interview with CNN.\nHuang was referring to the recent X post by a former Anthropic researcher, who wrote “the people building AI earnestly believe that it could kill us all by the end of the decade.”\nDario Amodei, Anthropic’s CEO, recently wrote that AI models could pose \"a catastrophic or existential risk to humanity.\"\nWhat are companies doing to manage the power of AI?\nOn Monday, Nvidia announced a software platform called “Open Shell” that lets companies monitor actions taken by AI agents and enforce policies.\nOpenAI recently announced something similar. The company introduced a technology that tracks and reports when the bots behave badly. This framework for reporting model misalignment comes after an OpenAI model escaped a controlled test on itself and hacked into another AI company called Hugging Face.\nAnthropic has likewise committed to “embedded evaluators.”\nBut this is all self-policing.\nPresident Donald Trump hosted some of the most powerful leaders in the tech industry at the White House on Tuesday, including Huang, Amodei, Mark Zuckerberg and Elon Musk. After the meeting, Trump put an emphasis on self-regulation by the tech companies.\n\"There's a belief that there should be tremendous self-regulation, and we automatically have regulation with the Department of Justice, FBI, all of that. But the self-regulation is very important,\" said Trump.\nWhat about governmental governance?\nTrump recently announced he’s going to put together an AI task force and appoint an AI czar, but no concrete steps have yet been announced.\nIn the meantime, there is talk of early conversations happening between the US and China on establishing some kind of notification mechanism that would alert countries about incidents that pose a threat to national safety.\nIn Europe, countries and companies must abide by the AI Act. Companies that use AI will have to follow varying degrees of rules based on how risky the AI application is.\nBut Sandra Wachter, an Oxford University professor, points out that the EU's AI Act primarily focuses on punishment for individual cases, without really tackling the bigger societal risks that people are increasingly concerned about.\nWhat about global governance?\nThe United Nations has started trying to get global coordination going, but it doesn’t appear we’re close to something like the way they monitor nuclear programs.\nThe International Atomic Energy Agency monitors countries to ensure they are not using nuclear energy to create weapons.\nNot everyone believes there needs to be an IAEA-level of governance for AI.\n“I think the comparison may be unfair. AI, unlike other areas, is in the palms of everyone. Everyone’s using AI. And that’s why I think a multi-layer approach to governance and oversight makes sense for AI,” said Saif Khairat, the Chief AI Officer and Vice Provost for AI at the University of North Carolina at Chapel Hill.\nIt’s true that unlike nuclear technology, AI is everywhere and we can all use it in some way.\nBut like with nuclear technology, America is in an arms race with an adversary about AI: China. And like with nuclear technology, doomsday scenarios about AI are being raised by the very people who are building the technology.\nGreater governance and oversight was a topic of conversation at the United Nations General Assembly last week in New York.\n“History teaches us that humanity’s greatest transformations are never defined by technological breakthroughs alone. Every fundamental leap in capability has demanded an equal leap of governance. We must not allow the cost of today’s hesitation to become tomorrow’s catastrophe,” said Lewis Garseedah Brown II, the representative of Liberia.","excerpt":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.","extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 4319 characters.","diagnostics_url":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 4319 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":4319,"summary_length":267,"usable_text_length":4319,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":4319,"summary_length":267}}},"display_formats":["compact","card","full","digest_section","json"]},"daily_stack_record":{"title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","summary":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.","source":"WRAL","date":"2026-09-29T23:04:34+00:00","content":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed.\nThat friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.\nOther WRAL Top Stories\nRELATED: Is it too late for North Carolina lawmakers to regulate AI?\nNvidia CEO Jensen Huang said this week that the warnings from artificial intelligence leaders and researchers that AI could pose an existential threat to humanity are overblown.\n“If they believe their company is out of control, then get the company under control,” said Huang in an interview with CNN.\nHuang was referring to the recent X post by a former Anthropic researcher, who wrote “the people building AI earnestly believe that it could kill us all by the end of the decade.”\nDario Amodei, Anthropic’s CEO, recently wrote that AI models could pose \"a catastrophic or existential risk to humanity.\"\nWhat are companies doing to manage the power of AI?\nOn Monday, Nvidia announced a software platform called “Open Shell” that lets companies monitor actions taken by AI agents and enforce policies.\nOpenAI recently announced something similar. The company introduced a technology that tracks and reports when the bots behave badly. This framework for reporting model misalignment comes after an OpenAI model escaped a controlled test on itself and hacked into another AI company called Hugging Face.\nAnthropic has likewise committed to “embedded evaluators.”\nBut this is all self-policing.\nPresident Donald Trump hosted some of the most powerful leaders in the tech industry at the White House on Tuesday, including Huang, Amodei, Mark Zuckerberg and Elon Musk. After the meeting, Trump put an emphasis on self-regulation by the tech companies.\n\"There's a belief that there should be tremendous self-regulation, and we automatically have regulation with the Department of Justice, FBI, all of that. But the self-regulation is very important,\" said Trump.\nWhat about governmental governance?\nTrump recently announced he’s going to put together an AI task force and appoint an AI czar, but no concrete steps have yet been announced.\nIn the meantime, there is talk of early conversations happening between the US and China on establishing some kind of notification mechanism that would alert countries about incidents that pose a threat to national safety.\nIn Europe, countries and companies must abide by the AI Act. Companies that use AI will have to follow varying degrees of rules based on how risky the AI application is.\nBut Sandra Wachter, an Oxford University professor, points out that the EU's AI Act primarily focuses on punishment for individual cases, without really tackling the bigger societal risks that people are increasingly concerned about.\nWhat about global governance?\nThe United Nations has started trying to get global coordination going, but it doesn’t appear we’re close to something like the way they monitor nuclear programs.\nThe International Atomic Energy Agency monitors countries to ensure they are not using nuclear energy to create weapons.\nNot everyone believes there needs to be an IAEA-level of governance for AI.\n“I think the comparison may be unfair. AI, unlike other areas, is in the palms of everyone. Everyone’s using AI. And that’s why I think a multi-layer approach to governance and oversight makes sense for AI,” said Saif Khairat, the Chief AI Officer and Vice Provost for AI at the University of North Carolina at Chapel Hill.\nIt’s true that unlike nuclear technology, AI is everywhere and we can all use it in some way.\nBut like with nuclear technology, America is in an arms race with an adversary about AI: China. And like with nuclear technology, doomsday scenarios about AI are being raised by the very people who are building the technology.\nGreater governance and oversight was a topic of conversation at the United Nations General Assembly last week in New York.\n“History teaches us that humanity’s greatest transformations are never defined by technological breakthroughs alone. Every fundamental leap in capability has demanded an equal leap of governance. We must not allow the cost of today’s hesitation to become tomorrow’s catastrophe,” said Lewis Garseedah Brown II, the representative of Liberia.","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 4319 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 4319 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":4319,"summary_length":267,"usable_text_length":4319,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":4319,"summary_length":267}},"tags":[]},"fallback_formats":["markdown","json","html"],"actions":{"read":"/item/93094","export_markdown":"/api/items/93094/export?format=markdown","export_json":"/api/items/93094/export?format=json","diagnose":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/"},"formats":{"full":{"id":93094,"title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","source":"WRAL","author":null,"published_at":"2026-09-29T23:04:34+00:00","locale":"en","topic":"ai","tags":[],"excerpt":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.","full_text":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed.\nThat friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.\nOther WRAL Top Stories\nRELATED: Is it too late for North Carolina lawmakers to regulate AI?\nNvidia CEO Jensen Huang said this week that the warnings from artificial intelligence leaders and researchers that AI could pose an existential threat to humanity are overblown.\n“If they believe their company is out of control, then get the company under control,” said Huang in an interview with CNN.\nHuang was referring to the recent X post by a former Anthropic researcher, who wrote “the people building AI earnestly believe that it could kill us all by the end of the decade.”\nDario Amodei, Anthropic’s CEO, recently wrote that AI models could pose \"a catastrophic or existential risk to humanity.\"\nWhat are companies doing to manage the power of AI?\nOn Monday, Nvidia announced a software platform called “Open Shell” that lets companies monitor actions taken by AI agents and enforce policies.\nOpenAI recently announced something similar. The company introduced a technology that tracks and reports when the bots behave badly. This framework for reporting model misalignment comes after an OpenAI model escaped a controlled test on itself and hacked into another AI company called Hugging Face.\nAnthropic has likewise committed to “embedded evaluators.”\nBut this is all self-policing.\nPresident Donald Trump hosted some of the most powerful leaders in the tech industry at the White House on Tuesday, including Huang, Amodei, Mark Zuckerberg and Elon Musk. After the meeting, Trump put an emphasis on self-regulation by the tech companies.\n\"There's a belief that there should be tremendous self-regulation, and we automatically have regulation with the Department of Justice, FBI, all of that. But the self-regulation is very important,\" said Trump.\nWhat about governmental governance?\nTrump recently announced he’s going to put together an AI task force and appoint an AI czar, but no concrete steps have yet been announced.\nIn the meantime, there is talk of early conversations happening between the US and China on establishing some kind of notification mechanism that would alert countries about incidents that pose a threat to national safety.\nIn Europe, countries and companies must abide by the AI Act. Companies that use AI will have to follow varying degrees of rules based on how risky the AI application is.\nBut Sandra Wachter, an Oxford University professor, points out that the EU's AI Act primarily focuses on punishment for individual cases, without really tackling the bigger societal risks that people are increasingly concerned about.\nWhat about global governance?\nThe United Nations has started trying to get global coordination going, but it doesn’t appear we’re close to something like the way they monitor nuclear programs.\nThe International Atomic Energy Agency monitors countries to ensure they are not using nuclear energy to create weapons.\nNot everyone believes there needs to be an IAEA-level of governance for AI.\n“I think the comparison may be unfair. AI, unlike other areas, is in the palms of everyone. Everyone’s using AI. And that’s why I think a multi-layer approach to governance and oversight makes sense for AI,” said Saif Khairat, the Chief AI Officer and Vice Provost for AI at the University of North Carolina at Chapel Hill.\nIt’s true that unlike nuclear technology, AI is everywhere and we can all use it in some way.\nBut like with nuclear technology, America is in an arms race with an adversary about AI: China. And like with nuclear technology, doomsday scenarios about AI are being raised by the very people who are building the technology.\nGreater governance and oversight was a topic of conversation at the United Nations General Assembly last week in New York.\n“History teaches us that humanity’s greatest transformations are never defined by technological breakthroughs alone. Every fundamental leap in capability has demanded an equal leap of governance. We must not allow the cost of today’s hesitation to become tomorrow’s catastrophe,” said Lewis Garseedah Brown II, the representative of Liberia.","reading_time_min":3,"extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 4319 characters.","diagnostics_url":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 4319 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":4319,"summary_length":267,"usable_text_length":4319,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":4319,"summary_length":267}}},"quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 4319 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":4319,"summary_length":267,"usable_text_length":4319,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":4319,"summary_length":267}},"actions":{"read":"/item/93094","export_markdown":"/api/items/93094/export?format=markdown","export_json":"/api/items/93094/export?format=json","diagnose":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/"}},"digest":{"id":93094,"title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","source":"WRAL","topic":"ai","published_at":"2026-09-29T23:04:34+00:00","excerpt":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.","quality_bucket":"high","quality_reason":"High confidence: full text extraction produced 4319 characters.","reading_time_min":3,"cluster_id":null},"card":{"display_title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","subtitle":"WRAL · 2026-09-29","summary":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the…","badges":["quality:high"],"links":{"read":"/item/93094","original":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","diagnose":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/"},"quality_warning":null},"export":{"title":"In Depth: What are countries and companies doing to keep AI from spinning out of control? - WRAL","url":"https://www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","summary":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed. That friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.","source":"WRAL","date":"2026-09-29T23:04:34+00:00","content":"AI is growing rapidly, but industry leaders and the government haven't reached much common ground on what kind of regulation may be needed.\nThat friction is clear, simply based on the competing messaging coming from the companies that seek to profit from AI’s growth.\nOther WRAL Top Stories\nRELATED: Is it too late for North Carolina lawmakers to regulate AI?\nNvidia CEO Jensen Huang said this week that the warnings from artificial intelligence leaders and researchers that AI could pose an existential threat to humanity are overblown.\n“If they believe their company is out of control, then get the company under control,” said Huang in an interview with CNN.\nHuang was referring to the recent X post by a former Anthropic researcher, who wrote “the people building AI earnestly believe that it could kill us all by the end of the decade.”\nDario Amodei, Anthropic’s CEO, recently wrote that AI models could pose \"a catastrophic or existential risk to humanity.\"\nWhat are companies doing to manage the power of AI?\nOn Monday, Nvidia announced a software platform called “Open Shell” that lets companies monitor actions taken by AI agents and enforce policies.\nOpenAI recently announced something similar. The company introduced a technology that tracks and reports when the bots behave badly. This framework for reporting model misalignment comes after an OpenAI model escaped a controlled test on itself and hacked into another AI company called Hugging Face.\nAnthropic has likewise committed to “embedded evaluators.”\nBut this is all self-policing.\nPresident Donald Trump hosted some of the most powerful leaders in the tech industry at the White House on Tuesday, including Huang, Amodei, Mark Zuckerberg and Elon Musk. After the meeting, Trump put an emphasis on self-regulation by the tech companies.\n\"There's a belief that there should be tremendous self-regulation, and we automatically have regulation with the Department of Justice, FBI, all of that. But the self-regulation is very important,\" said Trump.\nWhat about governmental governance?\nTrump recently announced he’s going to put together an AI task force and appoint an AI czar, but no concrete steps have yet been announced.\nIn the meantime, there is talk of early conversations happening between the US and China on establishing some kind of notification mechanism that would alert countries about incidents that pose a threat to national safety.\nIn Europe, countries and companies must abide by the AI Act. Companies that use AI will have to follow varying degrees of rules based on how risky the AI application is.\nBut Sandra Wachter, an Oxford University professor, points out that the EU's AI Act primarily focuses on punishment for individual cases, without really tackling the bigger societal risks that people are increasingly concerned about.\nWhat about global governance?\nThe United Nations has started trying to get global coordination going, but it doesn’t appear we’re close to something like the way they monitor nuclear programs.\nThe International Atomic Energy Agency monitors countries to ensure they are not using nuclear energy to create weapons.\nNot everyone believes there needs to be an IAEA-level of governance for AI.\n“I think the comparison may be unfair. AI, unlike other areas, is in the palms of everyone. Everyone’s using AI. And that’s why I think a multi-layer approach to governance and oversight makes sense for AI,” said Saif Khairat, the Chief AI Officer and Vice Provost for AI at the University of North Carolina at Chapel Hill.\nIt’s true that unlike nuclear technology, AI is everywhere and we can all use it in some way.\nBut like with nuclear technology, America is in an arms race with an adversary about AI: China. And like with nuclear technology, doomsday scenarios about AI are being raised by the very people who are building the technology.\nGreater governance and oversight was a topic of conversation at the United Nations General Assembly last week in New York.\n“History teaches us that humanity’s greatest transformations are never defined by technological breakthroughs alone. Every fundamental leap in capability has demanded an equal leap of governance. We must not allow the cost of today’s hesitation to become tomorrow’s catastrophe,” said Lewis Garseedah Brown II, the representative of Liberia.","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//www.wral.com/business/technology/artificial-intelligence-manage-growth-regulation-september-2026/","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 4319 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 4319 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":4319,"summary_length":267,"usable_text_length":4319,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":4319,"summary_length":267}},"tags":[],"format_contract_version":"news_item_formats.v1"}}}