HUGS 2027

1st International Workshop on Human-First Generative Software Engineering

Generative AI is becoming a lasting part of software engineering, reshaping how software-intensive systems are specified, designed, implemented, tested, documented, evolved, and operated. These capabilities promise faster and more powerful engineering workflows, but they also challenge established assumptions about human expertise, traceability, responsibility, independent assurance, certification, and societal trust.

HUGS 2027 invites researchers and practitioners to discuss how software engineering can ensure that GenAI-assisted development remains human-controlled, accountable, trustworthy, and aligned with ethical, legal, and societal values. The workshop welcomes contributions across software engineering, with particular attention to critical, high-impact, and software-intensive systems where responsibility, assurance, and trust are essential.

Call for Papers

1st International Workshop on Human-First Generative Software Engineering

Generative AI is becoming a lasting part of software engineering, reshaping how software-intensive systems are specified, designed, implemented, tested, documented, evolved, and operated. These capabilities promise faster and more powerful engineering workflows, but they also challenge established assumptions about human expertise, traceability, responsibility, independent assurance, certification, and societal trust.

HUGS 2027 invites researchers and practitioners to discuss how software engineering can ensure that GenAI-assisted development remains human-controlled, accountable, trustworthy, and aligned with ethical, legal, and societal values. The workshop welcomes contributions across software engineering, with particular attention to critical, high-impact, and software-intensive systems where responsibility, assurance, and trust are essential.

HUGS 2027 targets research and practices in AI for software engineering, software architecture, quality, model-driven engineering, requirements engineering, testing, verification and validation, DevOps/MLOps, assurance cases, empirical software engineering, human-AI interaction, software engineering for society, digitalization for sustainability (how SE can support sustainability by making systems within e.g. healthcare, energy, transport more efficient and safe), IoT and CPS engineering, and safety-critical systems.

Topics of interest include, but are not limited to:

• human-controlled GenAI for requirements, architecture, modeling, implementation, testing, documentation, maintenance, and operations;
• traceability, accountability, rationale, and provenance for AI-generated or AI-mediated engineering artifacts;
• human-centric assurance, independent review, human-in-the-loop verification and validation, and adaptive certification;
• transparent DevOps, MLOps, and TrustDevOps pipelines for GenAI-assisted software engineering;
• control surfaces, guardrails, prompts, workflows, and engineering interfaces for AI-assisted development;
• ethical assurance cases, value-sensitive engineering, governance models, and responsible AI integration in software engineering;
• empirical studies, industrial experience reports, benchmark proposals, tools, methods, and negative results;
• socio-technical, legal, regulatory, organizational, and policy challenges in GenAI-assisted software engineering.

Workshop format

HUGS is proposed as a one-day, interactive workshop at ICSE 2027 in Dublin, Ireland. Accepted papers will establish a shared basis through lightning talks, while moderated breakout groups will develop concrete research questions, risks, tool ideas, benchmark scenarios, and community actions. The exact workshop date is to be confirmed.

Submissions

HUGS solicits 3-page extended abstracts, including figures, tables, and references. Contributions may present early research ideas, position or vision papers, experience reports, industrial challenges, benchmark proposals, negative results, tool concepts, or research agendas. Accepted abstracts will be published in the IEEE workshop proceedings, and every submission will receive at least three reviews. The official publication date of the workshop proceedings is the date the proceedings are made available by IEEE. This date may be up to two weeks prior to the first day of ICSE 2027. The official publication date affects the deadline for any patent filings related to published work.

Anticipated outcomes

1. A shared research agenda for human-first generative software engineering.
2. Open problems and candidate benchmark scenarios for empirical and tool-building work.
3. A network of researchers and practitioners interested in continuing HUGS as a recurring venue.
4. A post-workshop report or article summarizing the roadmap developed by participants.

Proposed workshop focus

Three connected perspectives for a practical, human-first conversation.

Human-controlled design with GenAI

Engineering methods, environments, and control surfaces that let stakeholders define the role, boundaries, assumptions, and failure modes of GenAI in software-engineering workflows.

Human-centric assurance and oversight

Traceable AI-generated artifacts, transparent pipelines, independent review, adaptive certification, and evidence for accountable engineering decisions.

Ethical, legal, and societal values by design

Value-sensitive engineering, governance models, ethical assurance cases, and mechanisms that connect engineering practice to societal and normative commitments.