UK report lists potential AI risks and doomsday scenarios
The UK government released its “Capabilities and risks from frontier AI” report which will form the basis of discussion at…
DeepMind’s CEO draws comparison between AI risks and the climate crisis
Demis Hassabis, CEO of Google DeepMind, argued that AI’s risks should be taken as seriously as the climate crisis, and…
GPT-4’s multimodal capability makes it vulnerable to attack
The ability that GPT-4 has to process images is really impressive, but the new capability opens the model up to…
LLMs can infer personal data from your chat interactions
Companies like OpenAI and Meta say their models don’t collect personal data but their AI is really good at inferring…
China to attend the UK’s AI Summit at Bletchley Park
Despite recent controversies surrounding espionage allegations, China is set to participate in the upcoming UK AI Safety Summit at Bletchley…
Google executive emphasizes the importance of “getting AI right”
Matt Brittin, Google’s president for Europe, the Middle East, and Africa, emphasized that AI is “too important not to get…
SEC chair: AI will cause ‘unavoidable’ economic collapse
Chairman of the US Securities and Exchange Commission (SEC), Gary Gensler, says regulation of AI is essential to prevent a…
Researchers jailbreak GPT-4 using low-resource languages
Using low-resource languages (LRL) like Zulu or Scots Gaelic can elicit unsafe responses from GPT-4 despite its alignment guardrails. Researchers…
Drones are killing Russian soldiers autonomously in Ukraine
Drones boasting AI capabilities are now autonomously identifying and engaging targets in Ukraine. According to a report by the New…
Simply fine-tuning LLMs can remove alignment guardrails
Commercial LLMs like GPT-3.5 and OpenAI have guardrails to make sure the models are aligned and don’t generate dangerous responses.…