Critics take note: Khan Academy is on the move. While it’s not going to revolutionize education (or even computer science education), the Khan Academy’s new Computer Science program marks a great step forward for the Academy by making the lessons open-ended and tinker-ready. Okay! We get it! Here's a very thoughtful explanation by Minerva's Ari Bader-Natal about why the programming module really rocks-and how Khan can make it even better. Boy, did you let us have it! Dan Meyer, who has been sharp-tongued about Khan's math curriculum, lavished praise on Khan's programming modules, calling them 'really something special.' One or two readers came up to us last week at the Meetup and told us off. GPT-4 also shows no improvement over GPT-3.5 in some tests, including English language and art history exams.Editor's note: Last week, we at EdSurge were a tad dismissive of Khan Academy's newly launched computer science program. Results showed human judges preferred GPT-4 outputs over the most advanced variant of GPT-3.5 only about 61% of the time. One could argue GPT-4 represents only an incremental improvement over its predecessors in many practical scenarios.

That being whether alternative approaches are required to fundamentally solve issues of truthfulness and reliability, or whether throwing more data and resources at language models will eventually do the job. Its arrival will continue a crucial debate among critics. GPT-4 is much better at declining inappropriate requests and avoiding harmful content when compared to the initial ChatGPT release.

OpenAI used human feedback to fine-tune GPT-4 to produce more helpful and less problematic outputs. That said, OpenAI’s results on GPT-4 suggest it’s at least more reliable than previous GPT models. OpenAI has itself said GPT-4 is subject to the same limitations as previous language models, such as being prone to reasoning errors and biases, and making up false information. In other words, it’s not a new paradigm in AI research. Even though the GPT-4 technical report controversially provides no details about how the model was developed, all signs indicate it’s essentially a scaled-up version of GPT-3.5 with safety improvements.

Khan academy for coding