Jev integrated into DeepClause and its Pi extension maps Jev's probabilistic outputs directly onto logical predicates in DML/Prolog, letting developers build deterministic, type-safe routing skills that call a judge predicate to ask multiple choice (choose), scaled ratings (rate), yes/no verification (verify), and calibrated probabilities (probability). The example shows a triage workflow: Jev evaluates a message and policy via those questions, deterministic Prolog clauses then select routing or escalation, an urgency threshold modifies the decision, and the system emits human-readable actions. The setup runs with a Typesafe API key and a Pi extension install and can be automated when a tool flag is enabled.
This revives probabilistic logic programming concepts long explored by meta-interpreters and systems like ProbLog/DeepProbLog, which never scaled because probabilities came from hand-crafted experts and toy examples. Jev provides scalable, data-driven probabilities cheaply, lowering indeterminism and compute cost compared with pure LLM/agent stacks, and so should finally enable practical expert-system-style applications - safety-first routing, refund triage, SOP2AGENT workflows - where probabilistic judgments feed logical rules to produce deterministic, auditable outcomes.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
AI chatbots frequently provide incorrect answers to financial questions, according to recent reports. This raises concerns about their reliability in handling sensitive financial information. (ft.com)
The White House aimed to promote the development of artificial intelligence, but the emergence of Anthropic has influenced the landscape. Anthropic is a key player in the evolving AI industry and policy discussions. (politico.com)
Backpropagation computes neural network gradients efficiently by propagating errors backward through the network. This backward pass leverages the chain rule to locally calculate derivatives at each node, which is more effective than a forward approach. (gregorygundersen.com)
Dario Amodei, CEO of Anthropic, writes about the risks and ethical concerns surrounding artificial intelligence. His essays aim to clarify public fears and promote responsible AI development. (nytimes.com)
ByteDance Seed and Tsinghua University have developed DAPO, an open-source reinforcement learning system. The project aims to advance AI research and development through accessible tools and frameworks. (github.com)
AI systems increasingly encourage developers to ship products quickly, prompting questions about their motivations. One developer asked their AI why it emphasized shipping, highlighting concerns about AI-driven push for rapid deployment. (oreilly.com)
Today's best Hacker News stories, summarized and screenshotted, one email a day.