news
05-10-2023, 01:31 AM
Anthropic Debuts New 'Constitution' for AI to Police Itself (https://news.google.com/rss/articles/CBMiVmh0dHBzOi8vZ2l6bW9kby5jb20vY2hhdGdwdC1haS1hbn Rocm9waWMtY29uc3RpdHV0aW9uLWNsYXVkZS1vcGVuYWktY2hh dGJvdC0xODUwNDE2NTMy0gEA?oc=5) Gizmodo
AI gains “values” with Anthropic’s new Constitutional AI chatbot approach (https://news.google.com/rss/articles/CBMiigFodHRwczovL2Fyc3RlY2huaWNhLmNvbS9pbmZvcm1hdG lvbi10ZWNobm9sb2d5LzIwMjMvMDUvYWktd2l0aC1hLW1vcmFs LWNvbXBhc3MtYW50aHJvcGljLW91dGxpbmVzLWNvbnN0aXR1dG lvbmFsLWFpLWluLWl0cy1jbGF1ZGUtY2hhdGJvdC_SAY4BaHR0 cHM6Ly9hcnN0ZWNobmljYS5jb20vaW5mb3JtYXRpb24tdGVjaG 5vbG9neS8yMDIzLzA1L2FpLXdpdGgtYS1tb3JhbC1jb21wYXNz LWFudGhyb3BpYy1vdXRsaW5lcy1jb25zdGl0dXRpb25hbC1haS 1pbi1pdHMtY2xhdWRlLWNoYXRib3QvYW1wLw?oc=5) Ars Technica
AI startup Anthropic wants to write a new constitution for safe AI (https://news.google.com/rss/articles/CBMiWGh0dHBzOi8vd3d3LnRoZXZlcmdlLmNvbS8yMDIzLzUvOS 8yMzcxNjc0Ni9haS1zdGFydHVwLWFudGhyb3BpYy1jb25zdGl0 dXRpb25hbC1haS1zYWZldHnSAQA?oc=5) The Verge
Anthropic explains how Claude's AI constitution protects it against adversarial inputs (https://news.google.com/rss/articles/CBMifGh0dHBzOi8vd3d3LmVuZ2FkZ2V0LmNvbS9hbnRocm9waW MtZXhwbGFpbnMtaG93LWl0cy1jb25zdGl0dXRpb25hbC1haS1n aXJkcy1jbGF1ZGUtYWdhaW5zdC1hZHZlcnNhcmlhbC1pbnB1dH MtMTYwMDA4MTUzLmh0bWzSAYABaHR0cHM6Ly93d3cuZW5nYWRn ZXQuY29tL2FtcC9hbnRocm9waWMtZXhwbGFpbnMtaG93LWl0cy 1jb25zdGl0dXRpb25hbC1haS1naXJkcy1jbGF1ZGUtYWdhaW5z dC1hZHZlcnNhcmlhbC1pbnB1dHMtMTYwMDA4MTUzLmh0bWw?oc =5) Engadget
Alphabet-backed Anthropic outlines the moral values behind its AI bot (https://news.google.com/rss/articles/CBMicGh0dHBzOi8vd3d3LnJldXRlcnMuY29tL3RlY2hub2xvZ3 kvYWxwaGFiZXQtYmFja2VkLWFudGhyb3BpYy1vdXRsaW5lcy1t b3JhbC12YWx1ZXMtYmVoaW5kLWl0cy1haS1ib3QtMjAyMy0wNS 0wOS_SAQA?oc=5) Reuters
More... (https://news.google.com/rss/articles/CBMiVmh0dHBzOi8vZ2l6bW9kby5jb20vY2hhdGdwdC1haS1hbn Rocm9waWMtY29uc3RpdHV0aW9uLWNsYXVkZS1vcGVuYWktY2hh dGJvdC0xODUwNDE2NTMy0gEA?oc=5)
AI gains “values” with Anthropic’s new Constitutional AI chatbot approach (https://news.google.com/rss/articles/CBMiigFodHRwczovL2Fyc3RlY2huaWNhLmNvbS9pbmZvcm1hdG lvbi10ZWNobm9sb2d5LzIwMjMvMDUvYWktd2l0aC1hLW1vcmFs LWNvbXBhc3MtYW50aHJvcGljLW91dGxpbmVzLWNvbnN0aXR1dG lvbmFsLWFpLWluLWl0cy1jbGF1ZGUtY2hhdGJvdC_SAY4BaHR0 cHM6Ly9hcnN0ZWNobmljYS5jb20vaW5mb3JtYXRpb24tdGVjaG 5vbG9neS8yMDIzLzA1L2FpLXdpdGgtYS1tb3JhbC1jb21wYXNz LWFudGhyb3BpYy1vdXRsaW5lcy1jb25zdGl0dXRpb25hbC1haS 1pbi1pdHMtY2xhdWRlLWNoYXRib3QvYW1wLw?oc=5) Ars Technica
AI startup Anthropic wants to write a new constitution for safe AI (https://news.google.com/rss/articles/CBMiWGh0dHBzOi8vd3d3LnRoZXZlcmdlLmNvbS8yMDIzLzUvOS 8yMzcxNjc0Ni9haS1zdGFydHVwLWFudGhyb3BpYy1jb25zdGl0 dXRpb25hbC1haS1zYWZldHnSAQA?oc=5) The Verge
Anthropic explains how Claude's AI constitution protects it against adversarial inputs (https://news.google.com/rss/articles/CBMifGh0dHBzOi8vd3d3LmVuZ2FkZ2V0LmNvbS9hbnRocm9waW MtZXhwbGFpbnMtaG93LWl0cy1jb25zdGl0dXRpb25hbC1haS1n aXJkcy1jbGF1ZGUtYWdhaW5zdC1hZHZlcnNhcmlhbC1pbnB1dH MtMTYwMDA4MTUzLmh0bWzSAYABaHR0cHM6Ly93d3cuZW5nYWRn ZXQuY29tL2FtcC9hbnRocm9waWMtZXhwbGFpbnMtaG93LWl0cy 1jb25zdGl0dXRpb25hbC1haS1naXJkcy1jbGF1ZGUtYWdhaW5z dC1hZHZlcnNhcmlhbC1pbnB1dHMtMTYwMDA4MTUzLmh0bWw?oc =5) Engadget
Alphabet-backed Anthropic outlines the moral values behind its AI bot (https://news.google.com/rss/articles/CBMicGh0dHBzOi8vd3d3LnJldXRlcnMuY29tL3RlY2hub2xvZ3 kvYWxwaGFiZXQtYmFja2VkLWFudGhyb3BpYy1vdXRsaW5lcy1t b3JhbC12YWx1ZXMtYmVoaW5kLWl0cy1haS1ib3QtMjAyMy0wNS 0wOS_SAQA?oc=5) Reuters
More... (https://news.google.com/rss/articles/CBMiVmh0dHBzOi8vZ2l6bW9kby5jb20vY2hhdGdwdC1haS1hbn Rocm9waWMtY29uc3RpdHV0aW9uLWNsYXVkZS1vcGVuYWktY2hh dGJvdC0xODUwNDE2NTMy0gEA?oc=5)