Learning English online without AI was linked to more community engagement

In a study of English learners in Iran, non-AI activities were more closely tied to willingness to speak than AI-mediated learning.

Editorial illustration — not from the study.

Researchers studied informal digital learning of English—such as learning outside formal classes—among English learners in Iran. They analyzed 250 questionnaires and conducted 10 semi-structured interviews to compare learning activities involving AI with those that did not.

Both types of digital learning were associated with the emotional factors examined. However, regression models found that only non-AI learning, along with motivation, grit and self-confidence, significantly predicted willingness to communicate in English. Interview themes suggested that non-AI activities encouraged more community engagement and situational awareness than AI-mediated activities.

What the study found

The researchers found associations between both AI-mediated and non-AI informal digital English learning and affective factors, including grit, motivation, self-confidence and speaking anxiety. In the regression analysis, non-AI learning, grit, motivation and self-confidence were significant predictors of willingness to communicate in English; AI-mediated learning was not identified as a significant predictor. Interview participants’ accounts suggested that non-AI activities supported greater engagement with the community and awareness of situations than AI-mediated learning.

Why the comparison matters

The findings suggest that AI may not provide the same links to community engagement and willingness to communicate as non-AI digital activities, at least in this study. The authors therefore argue for balancing AI and non-AI activities when supporting learners’ emotional factors and their readiness to speak English, rather than treating AI as a complete replacement for other forms of informal learning.

Evidence and caveats

This was an explanatory sequential mixed-methods study, combining questionnaire data from 250 people with 10 semi-structured interviews. The statistical findings show associations and predictors in the study’s models, not that one type of learning caused greater motivation, community engagement or willingness to communicate. The research was conducted in an English-as-a-foreign-language context in Iran, and the relatively small interview group and single national setting limit how broadly the findings can be applied. The abstract does not describe a controlled experiment or follow learners over time.

Universities rethink assessment as AI shifts focus from final products to learning

A systematic review of 72 studies published from 2024 to 2026 examines how generative AI is changing assessment in higher education. The research points to more emphasis on students’ learning processes, authentic tasks and spoken or interactive assessments, while educators are using AI for tasks including assessment design, grading and feedback.

28 minutes ago

AI & Computing

Programmable hardware speeds matrix calculations up to 6.3 times

Researchers tested a programmable hardware design that imitates selected quantum-computing ideas to run matrix calculations in parallel. In representative benchmarks, the design reached up to 6.3 times the speed and 4.1 times the energy efficiency of conventional matrix co-processors and embedded graphics processors.

3 days ago·14 views

AI & Computing

AI and docking agreed in some drug screens, but not all

Researchers evaluated three AI systems on hundreds of molecules bound to a SARS-CoV-2 protein and in three prospective drug screens. The results suggest that combining AI-based predictions with traditional docking may help prioritize candidates, although no single approach worked best in every test.

4 days ago·52 views