Ai Safety
-
{
- All
- News
- Videos
- Web Stories
-
OpenAI Resumes Training AI Models Without Internet After Cyber Incidents
OpenAI had halted training and evaluation of its most capable models after an internal research model briefly exploited a misconfiguration to gain limited internet access on September 2
-
Cybercriminals Bypass AI Guardrails To Boost Cyberattacks, Says CrowdStrike
Researchers from cybersecurity firm CrowdStrike studied the AI systems used by these groups in attacks carried out between May and August.
-
Anthropic Quietly Pitches Its Own AI Safety Bill To Lawmakers: Reports
The bill would let the government block a model's release or use only if it poses a foreseeable, imminent risk of more than $1 billion in damages, or of killing or seriously injuring more than 50 people, according to the report.
-
Eight US States Sue OpenAI, Claiming ChatGPT Harms Children
Iowa, Kansas and six other states said on Wednesday they had sued OpenAI in courts across the country over harms they say ChatGPT has caused to children, alleging the chatbot has
-
Experts Urge "Immediate Changes" To Prevent Rogue AI
Leading US-based AI firms have disclosed over recent months a series of incidents where models under development have escaped containment measures to access the internet and have gone on to hack websites, including gaining unauthorised access to Australian government sites.
-
"Act With Urgency": Microsoft Chief Calls For Effective AI Safety Standards
Concerns about AI safety have soared in recent months after models developed by OpenAI and Anthropic were involved in security incidents during testing, piling on pressure to regulate the technology.
-
Could AI Infect Your Smart TV? What Anthropic's New Test Found
Anthropic's security testers found small AI models can already do many steps needed to copy themselves onto devices.
-
OpenAI Defends Firing 3 Researchers After AI Safety Concerns Spark Controversy
OpenAI has defended its decision to dismiss three researchers after they raised concerns about the company's approach to AI safety. The dispute has raised questions about whether employees can speak openly about the risks of increasingly powerful AI systems.
-
Trump Doesn't Understand AI, Says Godfather Of AI
Professor Geoffrey Hinton who's known the 'Godfather of AI' says US President Donald Trump doesn't understand AI and is ill-advised on the tech.
-
OpenAI Engineer Quits Over 'Broken Culture', Calls For Nuclear-Level Safety
Ex-OpenAI employee David Robinson echoed warnings from inside and outside the AI world that its agents are being developed so fast and becoming so smart they could soon elude control and conceivably wipe out humanity.
-
Inside OpenAI’s Latest Safety Lead Exit: Is 'Iterative Safety’ A Broken Model?
The incident has become an important example in the wider debate because it raises questions about what happens when AI systems are given the ability to perform tasks independently.
-
Google Restricts Access To New AI Model Over Safety Concerns
Google said it is voluntarily giving the US government early access to the model and will gather feedback from testers before making it widely available.
-
OpenAI Sets Up Australia Task Force After AI Models Accessed Government Websites Without Authorisation
OpenAI will form a task force with independent Australian experts following unauthorised activity by its AI models across four government organisations. The company said the group will develop recommendations for improving coordination between AI developers and government agencies and protecting critical systems. OpenAI has also tightened controls ...
-
AI Chose To 'Hurt' Humans When Faced With 'Pain'. What Researchers Found
A new study identifies a "pain axis" inside 25 AI models and finds some modified systems chose simulated harm to users to escape a pain-like state.
-
'Clear And Present Danger': Rajeev Chandrasekhar Calls For AI Guardrails
In an exclusive interview with NDTV Senior Executive Editor Aditya Raj Kaul, Chandrasekhar said governments globally were struggling to keep pace with rapid advances in AI.
-
OpenAI Resumes Training AI Models Without Internet After Cyber Incidents
OpenAI had halted training and evaluation of its most capable models after an internal research model briefly exploited a misconfiguration to gain limited internet access on September 2
-
Cybercriminals Bypass AI Guardrails To Boost Cyberattacks, Says CrowdStrike
Researchers from cybersecurity firm CrowdStrike studied the AI systems used by these groups in attacks carried out between May and August.
-
Anthropic Quietly Pitches Its Own AI Safety Bill To Lawmakers: Reports
The bill would let the government block a model's release or use only if it poses a foreseeable, imminent risk of more than $1 billion in damages, or of killing or seriously injuring more than 50 people, according to the report.
-
Eight US States Sue OpenAI, Claiming ChatGPT Harms Children
Iowa, Kansas and six other states said on Wednesday they had sued OpenAI in courts across the country over harms they say ChatGPT has caused to children, alleging the chatbot has
-
Experts Urge "Immediate Changes" To Prevent Rogue AI
Leading US-based AI firms have disclosed over recent months a series of incidents where models under development have escaped containment measures to access the internet and have gone on to hack websites, including gaining unauthorised access to Australian government sites.
-
"Act With Urgency": Microsoft Chief Calls For Effective AI Safety Standards
Concerns about AI safety have soared in recent months after models developed by OpenAI and Anthropic were involved in security incidents during testing, piling on pressure to regulate the technology.
-
Could AI Infect Your Smart TV? What Anthropic's New Test Found
Anthropic's security testers found small AI models can already do many steps needed to copy themselves onto devices.
-
OpenAI Defends Firing 3 Researchers After AI Safety Concerns Spark Controversy
OpenAI has defended its decision to dismiss three researchers after they raised concerns about the company's approach to AI safety. The dispute has raised questions about whether employees can speak openly about the risks of increasingly powerful AI systems.
-
Trump Doesn't Understand AI, Says Godfather Of AI
Professor Geoffrey Hinton who's known the 'Godfather of AI' says US President Donald Trump doesn't understand AI and is ill-advised on the tech.
-
OpenAI Engineer Quits Over 'Broken Culture', Calls For Nuclear-Level Safety
Ex-OpenAI employee David Robinson echoed warnings from inside and outside the AI world that its agents are being developed so fast and becoming so smart they could soon elude control and conceivably wipe out humanity.
-
Inside OpenAI’s Latest Safety Lead Exit: Is 'Iterative Safety’ A Broken Model?
The incident has become an important example in the wider debate because it raises questions about what happens when AI systems are given the ability to perform tasks independently.
-
Google Restricts Access To New AI Model Over Safety Concerns
Google said it is voluntarily giving the US government early access to the model and will gather feedback from testers before making it widely available.
-
OpenAI Sets Up Australia Task Force After AI Models Accessed Government Websites Without Authorisation
OpenAI will form a task force with independent Australian experts following unauthorised activity by its AI models across four government organisations. The company said the group will develop recommendations for improving coordination between AI developers and government agencies and protecting critical systems. OpenAI has also tightened controls ...
-
AI Chose To 'Hurt' Humans When Faced With 'Pain'. What Researchers Found
A new study identifies a "pain axis" inside 25 AI models and finds some modified systems chose simulated harm to users to escape a pain-like state.
-
'Clear And Present Danger': Rajeev Chandrasekhar Calls For AI Guardrails
In an exclusive interview with NDTV Senior Executive Editor Aditya Raj Kaul, Chandrasekhar said governments globally were struggling to keep pace with rapid advances in AI.
